阅读说明：本技术 相位恢复优化装置与相位恢复优化方法 (Phase recovery optimization device and phase recovery optimization method ) 是由 卓庭楠 郑凯文 童泰来 于 2018-06-26 设计创作，主要内容包括：一种相位恢复优化装置,适用于一相位恢复电路中的一锁相回路,其中该锁相回路包含一回路滤波器、一振荡器与一相位估测器,其中该相位恢复优化装置包含一相位旋转器,根据分别自该相位估测器与该振荡器所接收的一估计相位与一补偿相位产生一补偿后相位；一变异数计算单元,根据该补偿后相位计算一补偿后相位变异数；以及一增益系数优化单元,根据该补偿后相位变异数调整该回路滤波器的增益系数。(A phase recovery optimization device is suitable for a phase-locked loop in a phase recovery circuit, wherein the phase-locked loop comprises a loop filter, an oscillator and a phase estimator, wherein the phase recovery optimization device comprises a phase rotator for generating a compensated phase according to an estimated phase and a compensated phase received from the phase estimator and the oscillator respectively; a variance calculating unit for calculating a compensated phase variance according to the compensated phase; and a gain coefficient optimizing unit for adjusting the gain coefficient of the loop filter according to the compensated phase variation.)

1. A phase recovery optimization apparatus for a phase locked loop in a phase recovery circuit, wherein the phase locked loop comprises a loop filter, an oscillator and a phase estimator, wherein the phase recovery optimization apparatus comprises:

a phase rotator for generating a compensated phase based on an estimated phase and a compensated phase received from the phase estimator and the oscillator, respectively;

a variance calculating unit for calculating a compensated phase variance according to the compensated phase; and

and a gain coefficient optimizing unit for adjusting a gain coefficient of the loop filter according to the compensated phase variance.

2. The phase recovery optimization device of claim 1, wherein the gain factor optimization unit is configured to:

adjusting the gain factor towards a trend; and

comparing a current compensated phase variation number with a previous compensated phase variation number to generate a judgment result, wherein the current compensated phase variation number corresponds to a gain coefficient after adjustment, and the previous compensated phase variation number corresponds to a gain coefficient before adjustment.

3. The phase recovery optimization device of claim 2, wherein the gain factor optimization unit is further configured to:

adjusting the gain factor toward an opposite trend according to the determination result indicating that the current compensated phase variation is greater than the previous compensated phase variation.

4. The phase recovery optimization device of claim 2, wherein the gain factor optimization unit is further configured to:

and canceling the previous adjustment of the gain factor according to the judgment result indicating that the current compensated phase variation is larger than the previous compensated phase variation.

5. The phase recovery optimization device of claim 2, wherein the gain factor optimization unit is further configured to:

adjusting the gain factor toward the trend according to the determination result indicating that the current compensated phase variation is smaller than the previous compensated phase variation.

6. The phase recovery optimization device of claim 1, wherein the gain factor optimization unit is further configured to:

storing a gain coefficient before adjustment; and

and judging whether the gain coefficient is adjusted or not according to the gain coefficient before adjustment.

7. The phase recovery optimization device of claim 6, wherein the gain factor optimization unit is further configured to:

by comparing the gain coefficient before adjustment with a gain coefficient before adjustment, whether the adjustment of the gain coefficient is completed is judged.

8. A phase recovery optimization method is applied to a phase-locked loop in a phase recovery circuit, wherein the phase-locked loop comprises a loop filter, an oscillator and a phase estimator, and the phase recovery optimization method comprises the following steps:

generating a compensated phase based on an estimated phase and a compensated phase received from the phase estimator and the oscillator, respectively;

calculating a compensated phase variation number according to the compensated phase; and

adjusting a gain coefficient of the loop filter according to the compensated phase variance.

9. The method of claim 8, wherein the step of adjusting the gain factor of the loop filter according to the compensated phase variance comprises:

adjusting the gain factor towards a trend; and

comparing a current compensated phase variation number with a previous compensated phase variation number to generate a judgment result, wherein the current compensated phase variation number corresponds to a gain coefficient adjusted towards the trend, and the previous compensated phase variation number corresponds to a gain coefficient before adjustment.

10. The method of claim 9, wherein the step of adjusting the gain factor of the loop filter according to the compensated phase variance further comprises:

adjusting the gain factor toward an opposite trend according to the determination result indicating that the current compensated phase variation is greater than the previous compensated phase variation.

11. The method of claim 9, wherein the step of adjusting the gain factor of the loop filter according to the compensated phase variance further comprises:

and canceling the previous adjustment of the gain factor according to the judgment result indicating that the current compensated phase variation is larger than the previous compensated phase variation.

12. The method of claim 9, wherein the step of adjusting the gain factor of the loop filter according to the compensated phase variance further comprises:

and continuing to adjust the gain factor towards the trend according to the judgment result indicating that the current compensated phase variation is smaller than the previous compensated phase variation.

13. The method of claim 8, wherein the step of adjusting the gain factor of the loop filter according to the compensated phase variance further comprises:

storing a gain factor before adjustment; and

and judging whether the gain coefficient is adjusted or not according to the gain coefficient before adjustment.

14. The phase recovery optimization method of claim 13, wherein the step of determining whether the gain factor is adjusted according to the gain factor before the adjustment comprises:

by comparing the gain factor before adjustment with a gain factor after adjustment, it is determined whether the adjustment of the gain factor is completed.

Technical Field

The present invention relates to the field of communication systems, and more particularly, to a phase recovery optimization apparatus and a phase recovery optimization method.

Background

In a communication system, a transmitting end processes, modulates, filters, amplifies, and transmits data to one or more receiving ends, and before a transmitted signal reaches the receiving end, the transmitted signal is usually weakened by path loss (path loss), multipath interference (multipath interference), and other types of signals (signal degradation). The receiving end performs various types of adjustment on the transmission signal and demodulates the adjusted signal to recover the transmission data.

In the demodulation process, the receiving end often employs a phase recovery circuit to remove the phase change existing between the transmitting end oscillator and the receiving end oscillator, and the phase recovery circuit often combines a phase-locked loop (PLL) to track and remove the phase change. Generally, a phase-locked loop includes a loop filter (loop filter), wherein the gain coefficients of the loop filter are fixed values that are optimized according to an assumed phase change. However, the actual phase change is likely to be different from the assumed phase change, thus compromising the performance of the phase recovery circuit.

Disclosure of Invention

Therefore, an object of the present invention is to provide a phase recovery optimization apparatus and method, which can dynamically adjust the gain factor of the loop filter according to the actual phase change.

The invention discloses a phase recovery optimizing device, which is suitable for a phase-locked loop in a phase recovery circuit, wherein the phase-locked loop comprises a loop filter, an oscillator and a phase estimator, wherein the phase recovery optimizing device comprises a phase rotator which generates a compensated phase according to an estimated phase and a compensated phase which are respectively received from the phase estimator and the oscillator; a variance calculating unit for calculating a compensated phase variance according to the compensated phase; and a gain coefficient optimizing unit for adjusting the gain coefficient of the loop filter according to the compensated phase variation.

The present invention further discloses a phase recovery optimization method, which is applied to a phase locked loop in a phase recovery circuit, wherein the phase locked loop comprises a loop filter, an oscillator and a phase estimator, and the phase recovery optimization method comprises generating a compensated phase according to an estimated phase and a compensated phase received from the phase estimator and the oscillator, respectively; calculating a compensated phase variation number according to the compensated phase; and adjusting the gain coefficient of the loop filter according to the compensated phase variation.

The features, operation and function of the present invention will be described in detail with reference to the drawings.

Drawings

FIG. 1 shows a block diagram of a conventional phase recovery circuit;

FIG. 2 is a diagram illustrating an exemplary frame in accordance with one embodiment of the present invention;

FIG. 3 is a block diagram of a phase recovery circuit according to an embodiment of the present invention;

FIG. 4 is a detailed block diagram of a phase recovery optimization apparatus according to an embodiment of the present invention;

FIG. 5 is a flow chart illustrating adjusting the loop filter gain factor according to an embodiment of the present invention; and

fig. 6 is a detailed flowchart of step S520 according to an embodiment of the present invention.

Description of the symbols

100. 300 phase recovery circuit

110. 310 phase rotator

120. 320 phase error detector

130. 330 loop filter

140. 340 oscillator

150. 350 phase estimator

132. 134, 332, 334 multiplier

136. 336 accumulator

138. 338, 362 adders

360 phase recovery optimizing device

364 variance calculating unit

366 gain factor optimization unit

R signal

φ_OCompensating phase

R' compensated signal

φ_estEstimating phase

K_P、K_IGain factor

FRA frame

H header subframe

D0-Dn data subframes

P1 Pn pilot subframe

θ_MLMaximum likelihood phase

y_iKnown symbol

m_iKnown data

θ_ML' post-compensation maximum likelihood phase

post-V compensation phase variation

Vp phase variance after previous compensation

Vc current compensated phase variation

S510 to S540

Detailed Description

Fig. 1 is a block diagram of a conventional phase recovery circuit 100. The phase recovery circuit 100 includes a phase rotator 110, a phase error detector 120, a loop filter 130, an oscillator 140, and a phase estimator 150. The signal R processed by the previous stage circuit of the phase recovery circuit 100 is input to the phase recovery circuit 100 to estimate and compensate for a phase error in the signal R.

In detail, the signal R is input to the phase rotator 110 and has a compensated phase phi with the output of the oscillator 140_OThe phase rotator 110 may be implemented as a multiplier that mixes to generate a compensated signal R'. The phase error detector 120 generates an estimated phase phi according to the compensated signal R_estWhich is used as a phase tracking parameter and is input to the loop filter 130. The loop filter 130 includes multipliers 132, 134, an accumulator 136 and an adder 138, wherein the multiplier 132 is formed to have a gain coefficient K_PThe proportional filter (proportional filter), the multiplier 134 and the accumulator (accumulator)136 form a filter with a gain factor K_IIntegral filter (integrated filter). Estimating phase phi_estIs input in parallel to a proportional filter and an integral filter, the proportional filter uses a multiplier 132 to estimate the phase phi_estMultiplying by a gain factor K_P(ii) a In parallel, the integrator filter uses the multiplier 134 to estimate the phase phi_estMultiplying by a gain factor K_IThen the result K_I*φ_estTo the accumulator 136 for accumulation. The adder 138 then adds the outputs of the multiplier 132 and the accumulator 136 and provides the result to the oscillator 140, and the oscillator 140 then generates the compensation phase phi according to the output result of the adder 138_OIn practice, the oscillator 140 may be an NCO (numerically controlled oscillator). In one embodiment, the compensation phase is_OThe start value of (a) may be provided by the phase estimator 150. the phase estimator 150 may generate a Maximum Likelihood phase θ from known data (knock data) in a frame (frame) using a Maximum Likelihood (ML) estimation method_MLAs a compensation phase phi_OA starting value of (a). Referring to fig. 2, fig. 2 is a diagram of an exemplary frame FRA, where the frame FRA includes a header subframe H, a plurality of data subframes D0-Dn and a plurality of pilot subframes P1-Pn, each of the subframes includes a plurality of symbols (symbols), for example, the header subframe H may include 90 symbols and the pilot subframe P1 may include 36 symbols, where the header subframe H and the plurality of pilot subframes P1-Pn belong to known data, and thus the header subframe H and the plurality of pilot subframes P1-Pn can be used to calculate the actual phase error and the maximum likelihood phase θ_MLCan be represented by the following formula:

wherein, y_iRepresented as a known symbol (symbol), m, in the signal R_iThe known data representing the symbol. For example, the phase estimator 150 can utilize 90 known symbols of the header subframe H to generate a maximum likelihood phase θ_MLAs a compensation phase phi_OThen the phase rotator 110 will combine a symbol in the data sub-frame D0 with the compensated phase phi outputted from the oscillator 140_OMixing to generate a compensated signal R', the phase error detector 120 generating an estimated phase phi according to the compensated signal R_estEstimate the phase phi_estFiltered by the loop filter 130 and input to the oscillator 140 to generate an updated compensated phase phi_OFor compensating the next symbol in the data sub-frame D0. By analogy with thatAll symbols in the data sub-frame D0 are compensated.

However, the gain factor K of the loop filter 140_P、K_IUsually a fixed value, which cannot be adjusted to the actual channel conditions, compromising the performance of the phase recovery circuit. Therefore, the present invention dynamically adjusts the gain coefficient of the loop filter 140 through a phase recovery optimization device to improve the performance of the overall phase recovery circuit. FIG. 3 is a block diagram of a phase recovery circuit 300 according to an embodiment of the present invention, wherein the phase recovery circuit 300 further comprises a phase recovery optimization device 360 coupled to the oscillator 340 and the phase estimator 350 for receiving a compensated phase φ from the oscillator 340 and the phase estimator 350, respectively, in comparison with the conventional phase recovery 100_OAnd a maximum likelihood phase theta_MLWith the compensation phase phi_OCompensating for the maximum likelihood phase theta_MLTo generate a compensated maximum likelihood phase theta_MLAnd based on the compensated maximum likelihood phase theta_ML'A variance (variance) of the' is used to adjust the gain factor K of the loop filter 340_P、K_I。

In detail, taking fig. 2 as an example, it is assumed that the data sub-frame D0 is a gain coefficient k_p、k_iFor phase estimation and compensation, the oscillator 340 will generate the compensation phase phi according to the last symbol of the data sub-frame D0_OTransmitted to the phase recovery optimization device 360; in parallel, the phase estimator 350 will generate the maximum likelihood phase θ according to the 36 known symbols of the pilot subframe P1_MLTo the phase recovery optimization device 360. Referring to fig. 4, fig. 4 is a detailed block diagram of the phase recovery optimization apparatus 360. as shown in fig. 4, the phase recovery optimization apparatus 360 includes an adder 362, a variance calculation unit 364 and a gain factor optimization unit 366. The adder 362 in the phase recovery optimization device 360 utilizes the compensated phase phi_OCompensating for maximum likelihood phase θ_MLTo generate a compensated maximum likelihood phase theta_ML'. Similarly, data sub-frame D1 is a gain factor k_p、k_iFor phase estimation and compensation, the oscillator 340 will generate the compensation phase phi according to the last symbol of the data sub-frame D1_OTransmitted to the phase recovery optimization device 360; in parallel, the phase estimator 350 will generate the maximum likelihood phase θ according to the 36 known symbols of the pilot subframe P2_MLTransmitted to the phase recovery optimizing device 360, and the adder 362 in the phase recovery optimizing device 360 uses the compensated phase phi_OCompensating for maximum likelihood phase θ_MLTo generate another compensated maximum likelihood phase theta_ML'. By analogy, adder 362 outputs multiple compensated maximum likelihood phases θ_ML' the variance calculating unit 364 calculates the maximum likelihood phase θ based on the plurality of compensated maximum likelihood phases_ML' calculation corresponds to the gain factor k_p、k_iA compensated phase variation V. According to the desired gain factor K_P、K_IEstimated compensation phase phi_OWill approximate the maximum likelihood phase theta_MLSo using the compensation phase phi_OCompensating for maximum likelihood phase θ_MLA resulting compensated maximum likelihood phase θ_ML' will be approximately 0, so that the phase variation V after compensation is closer to 0 to represent its corresponding gain coefficient K_P、K_IThe better, the gain factor optimizing unit 366 can adjust the gain factor K according to the compensated phase variation V_P、K_I。

In detail, referring to fig. 5, fig. 5 is a flow chart of adjusting the gain factor. First, the gain factor optimizing unit 366 stores the gain factor before adjustment (step S510) as a basis for determining whether to continue adjusting the gain factor. For example, if the gain factor K is not adjusted_P＝k_p、K_I＝k_iThe gain factor optimizing unit 366 will select k_p、k_iAnd (4) storing. Then, the gain factor optimizing unit 366 first adjusts the gain factor K_P、K_IIn one of the above embodiments, the gain factor optimizing unit 366 first adjusts the gain factor K_P(step S520), the gain factor K is adjusted_I(step S530), in other embodiments, the gain factor optimizing unit 366 may also adjust the gain factor K first_IReadjusting the gain factor K_P。

FIG. 6 shows the stepsA detailed flowchart of step S520. First, the gain factor optimizing unit 366 adjusts the gain factor K toward a trend_P(step S521), for example, the gain factor optimizing unit 366 can increase the gain factor K_PLet K be_P＝k_p+1>k_pAnd corresponds to the gain coefficient k_p+1、k_iA compensated phase variation Vp +1 of (a) may be generated as described above. Next, the gain factor optimizing unit 366 compares a previous compensated phase variation Vp with a current compensated phase variation Vc (step S522) to determine whether to continue to adjust the gain factor K toward the trend_PAt this time, the previously compensated phase variation Vp is equal to V, which corresponds to the gain coefficient k before adjustment_p、k_i(ii) a The current post-compensation phase variation Vc ═ Vp +1, which corresponds to the adjusted gain coefficient k_p+1、k_i。

In step S522, if the current compensated phase variation Vc is smaller than the previous compensated phase variation Vp, it means that the adjusted gain factor is better, and therefore the gain factor optimizing unit 366 continues to adjust the gain factor K toward the trend_P(step S523), in this case, the gain factor optimizing unit 366 continues to increase the gain factor K_PLet K be_P＝k_p+2>k_p+1 and corresponds to the gain factor k_p+2、k_iA compensated phase variation Vp +2 of (a) may be generated as previously described. Next, the gain factor optimizing unit 366 compares a previous compensated phase variation Vp with a current compensated phase variation Vc (step S524) to determine whether to continue adjusting the gain factor K_PAt this time, the previous post-compensation phase variation Vp is Vp +1, which corresponds to the gain coefficient k before adjustment_p+1、k_i(ii) a The current post-compensation phase variation Vc ═ Vp +2, which corresponds to the adjusted gain coefficient k_p+2、k_i. In step S524, if the current compensated phase variation Vc is greater than the previous compensated phase variation Vp, which represents that the adjusted gain factor is poor, the gain factor optimizing unit 366 cancels the previous gain factor K_PIs adjusted (step S525) so that the gain coefficient K becomes equal to_PFrom k to k_p+2 to k_p+1, and end on the gain factor K_PIn other words the final gain factor K_P＝k_p+ 1; in step S524, if the current compensated phase variation Vc is smaller than the previous compensated phase variation Vp, which means that the adjusted gain factor is better, the gain factor optimizing unit 366 continues to adjust the gain factor K toward the trend_P(step S523), in this case, the gain factor optimizing unit 366 continues to increase the gain factor K_PLet K be_P＝k_p+3>k_p+2, ending the adjustment of the gain factor K until the phase variation Vc after the current compensation is larger than the phase variation Vp after the previous compensation_PUntil the adjustment.

In step S522, if the current compensated phase variation Vc is greater than the previous compensated phase variation Vp, which represents that the adjusted gain factor is poor, the gain factor optimizing unit 366 cancels the previous gain factor K_PAnd adjusting the gain factor K towards an opposite trend_P(step S526), in this case, the gain factor optimizing unit 366 decreases the gain factor K_PMake K_P＝k_p-1<k_pAnd corresponds to the gain coefficient k_p-1、k_iA compensated phase variation Vp-1 of (a) may be generated as previously described. Next, the gain factor optimizing unit 366 compares a previous compensated phase variation Vp with a current compensated phase variation Vc (step S527), where the previous compensated phase variation Vp equals V, which corresponds to the gain factor k before adjustment_p、k_i(ii) a The current post-compensation phase variation Vc ═ Vp-1, which corresponds to the adjusted gain coefficient k_p-1、k_i. In step S527, if the current compensated phase variation Vc is greater than the previous compensated phase variation Vp, which represents that the adjusted gain factor is poor, the gain factor optimizing unit 366 cancels the previous gain factor K_PAdjustment of (step S525), i.e. gain factor K_PFrom k to k_p-1 reverts to k_pAnd end on the gain factor K_PIn other words the final gain factor K_P＝k_p(ii) a In step S527, if the current compensated phase variation Vc is smallSince the previous compensated phase variation Vp represents a better adjusted gain factor, the gain factor optimization unit 366 continues to adjust the gain factor K toward the opposite trend_P(step S526), in this case, the gain factor optimizing unit 366 continues to decrease the gain factor K_PMake K_P＝k_p-2<k_p-1, ending the pair of gain coefficients K until the present compensated phase variation Vc is greater than the previous compensated phase variation Vp_PUntil the adjustment.

The gain factor optimizing unit 366 adjusts the gain factor K at the end_PAfter the adjustment (step S520), the gain coefficient K is then adjusted_I(step S530), since the adjustment methods of the two are similar, the related details can be referred to fig. 6, and are not described herein again.

In the pair of gain coefficients K_P、K_IAfter the adjustment is completed, the gain factor optimizing unit 366 then determines the gain factor K according to the pre-adjustment gain factor stored in step S510_P、K_IWhether the adjustment is completed or not (step S540) is determined to adjust the gain factor K again or not_P、K_I。

In one embodiment, the gain factor optimizing unit 366 can compare the gain factor K before and after adjustment_P、K_IWhether the gain coefficients are the same or not is judged_P、K_IWhether the adjustment is complete. If the gain factor K before and after adjustment_P、K_ISame, represents the gain coefficient K_P、K_IWhen the adjustment is completed, the gain factor optimizing unit 366 does not adjust the gain factor K_P、K_IThe adjustment is performed again, and then the phase recovery circuit 300 performs the adjustment with the gain factor K after the adjustment is completed_P、K_IEstimating and compensating for the phase error in the signal R until, for example, when the user switches channels, the gain factor optimization unit 366 re-aligns the gain factor K_P、K_IAnd (6) adjusting. With gain factor K before adjustment_P＝k_p、K_I＝k_iFor example, if the gain coefficient K is adjusted in steps S520 and S530_P＝k_p、K_I＝k_iCoefficient of gainThe optimization unit 366 does not match the gain factor K_P、K_IThe adjustment is performed again, and the phase recovery circuit 300 performs the adjustment with the gain factor K_P＝k_p、K_I＝k_iThe phase error in the signal R is estimated and compensated. If the gain factor K before and after adjustment_P、K_IDifferent from, representing the gain factor K_P、K_IIf the adjustment is not completed, the gain factor optimizing unit 366 will optimize the gain factor K_P、K_IAdjustment is performed again (return to step S510). With gain factor K before adjustment_P＝k_p、K_I＝k_iFor example, if the gain coefficient K is adjusted in steps S520 and S530_P＝k_p+1、K_I＝k_iThen the gain factor optimizing unit 366 would compare the gain factor K_P、K_IAdjustment is again made until the gain factor K_P、K_IThe adjustment of (2) is completed.

In another embodiment, the gain factor optimizing unit 366 can also determine the gain factor K before and after adjustment_P、K_IWhether the difference is less than a predetermined threshold value to determine the gain coefficient K_P、K_IWhether the adjustment is completed (step S540). If the gain factor K before and after adjustment_P、K_IIs smaller than the predetermined threshold value, representing a gain factor K_P、K_IFinishing the adjustment; on the contrary, it represents the gain coefficient K_P、K_IThe adjustment is not completed.

In another embodiment, if the gain factor K is_P、K_IThe adjustment is not completed (step S540), which means that the number of samples currently used for calculating the compensated phase variation V may be too small to effectively remove the noise, so that the phase recovery optimization apparatus 360 increases the number of samples for calculating the compensated phase variation V to match the gain factor K again_P、K_IAnd (6) adjusting.

In summary, the present invention dynamically adjusts the gain coefficient of the loop filter according to the compensated phase variation, so as to improve the performance of the overall phase recovery circuit.

The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes performed by the present specification and drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.