Decision feedback adaptive equalizer with low-frequency compensation unit and method

Abstract

The invention discloses a decision feedback self-adaptive equalizer with a low-frequency compensation unit and a method, and belongs to the technical field of digital communication, the decision feedback self-adaptive equalizer is provided with the low-frequency compensation unit, output signals of a plurality of multipliers are superposed by using an output signal b (k) of the low-frequency compensation unit, and the output signals b (k) of the low-frequency compensation unit are obtained. Compensation of multipath interference signals is realized; further, a judgment signal d (k) of the current moment k is generated by using an output signal y (k) of the current moment k of the adder, and an error signal e (k) of the current moment k is calculated by using a formula e (k) = d (k)-y (k), so that the filter coefficient and the low-frequency signal compensation amount of the next moment are updated, and the change of a channel is adapted; finally, the problem that a traditional decision feedback adaptive equalizer in an intensity modulation direct detection communication system cannot correctly equalize multipath interference signals is solved, and the operation stability of the communication system is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital communication, and more specifically, relates to a decision feedback adaptive equalizer with a low-frequency compensation unit and a method. Background Art

[0002] The intensity modulation direct detection optical transmission system has significant advantages in terms of cost and reliability, and has been widely used in traditional high-speed, long-distance optical communication systems. With the continuous growth of high-speed metropolitan and access optical transmission demand, the application potential of optical communication technology based on intensity modulation direct detection in metropolitan and access fields is more obvious.

[0003] The intensity modulation direct detection communication system is a fiber optic communication system that relies on light field intensity coding to transmit information. Intensity modulation direct detection is to directly perform envelope detection on the intensity modulated optical wireless signal. In other words, the intensity modulated signal can be directly restored to the original signal through a photodetector. It is a fiber optic communication system in which the transmitter modulates the intensity of the optical carrier with a signal and the receiver uses a detector to directly detect the optical signal.

[0004] Due to improper maintenance of the link connection points, the intensity modulation direct detection communication system will have serious multipath effects, which will cause serious signal amplitude imbalance, thereby increasing the system's bit error rate and reducing communication quality. Therefore, it is necessary to perform equalization operations on the received signal at the receiving end of the system to reduce or eliminate the impact of multipath interference. However, the traditional decision feedback adaptive equalizer in the intensity modulation direct detection communication system cannot correctly equalize the multipath interference signal. Summary of the invention

[0005] In view of the above defects or improvement needs of the prior art, the present invention provides a decision feedback adaptive equalizer and method with a low-frequency compensation unit, the purpose of which is to solve the technical problem that the traditional decision feedback adaptive equalizer in the current intensity modulation direct detection communication system cannot correctly equalize the multipath interference signal.

[0006] To achieve the above object, according to one aspect of the present invention, a decision feedback adaptive equalizer with a low frequency compensation unit is provided, comprising:

[0007] A plurality of sequentially connected delay units, wherein the first delay unit is used to delay the received signal x(k) of the decision feedback adaptive equalizer corresponding to the current time k and output x(k-1), and the i-th delay unit is used to delay the output signal of the i-1-th delay unit and output x(k-i+1); i∈[1, L-1], L-1 is the total number of the delay units;

[0008] A plurality of multipliers are connected to the plurality of delay units in a one-to-one correspondence, the input end of each multiplier is connected to the input end of the corresponding delay unit, and the i-th multiplier is used to multiply the input signal x(k-i+1) of the i-th delay unit by the i-th preset filter coefficient w at the current time k i-1 (k) multiply and output w i-1 (k)*x(k-i+1);

[0009] A low-frequency compensation unit, used to output a preset low-frequency signal b(k) at the current moment k;

[0010] an adder connected to the output ends of the plurality of multipliers and the output end of the low-frequency compensation unit, and configured to sum the output signals of the plurality of multipliers and the output signal of the low-frequency compensation unit and output y(k);

[0011] A decision device, connected to the output end of the adder, for generating a decision signal d(k) at the current time k using the output signal y(k) of the adder at the current time k;

[0012] A subtractor, connected to the output end of the adder and the output end of the decision device, for calculating the error signal e(k) at the current moment k using the formula e(k)=d(k)-y(k);

[0013] A controller connected to the subtractor, the plurality of multipliers and the low-frequency compensation unit, configured to update each preset filter coefficient w at the next moment k+1 using the error signal e(k) at the current moment k i-1 (k+1) and a preset low frequency signal b(k+1).

[0014] In one embodiment, the i-th preset filter coefficient w at the next time k+1 i-1 (k+1) is the error signal e(k) at the current time k and the i-th preset filter coefficient w at the current time k i-1 (k) Set.

[0015] In one embodiment, the formula w(k+1)=w(k)+μ w e(k)x * (k) Calculate the i-th preset filter coefficient w at the next moment k+1 i-1 (k+1), μ w is the update rate of the filter coefficients, x * (k) is the conjugate of x(k).

[0016] In one embodiment, the preset low frequency signal b(k+1) of the low frequency compensation unit at the next time k+1 is set by using the error signal e(k) at the current time k and the preset low frequency signal b(k) at the current time k.

[0017] In one embodiment, the formula b(k+1)=b(k)+μ b e(k) calculates the preset low-frequency signal b(k+1) at the next moment k+1; where μ b is the update rate of the low frequency compensation unit.

[0018] In one embodiment, the output of the adder is:

[0019] y(k)=[w0(k),w1(k),...,w L-1 (k)] * ×[x(k),x(k-1),...,x(k-L+1)] T +b(k);

[0020] The superscript * stands for conjugate and T stands for transpose.

[0021] According to another aspect of the present invention, a decision feedback adaptive equalization method with a low frequency compensation unit is provided, comprising:

[0022] S1: When the received signal x(k) is received at the current time k, a plurality of sequentially connected delay units are used to delay and output; the i-th delay unit is used to delay the output signal of the i-1-th delay unit and output x(k-i+1); i∈[1, L-1], L-1 is the total number of the delay units;

[0023] S2: Use each multiplier to multiply the input signal of the corresponding delay unit with the corresponding preset filter coefficient and output it. The i-th multiplier multiplies the input signal x(k-i+1) of the i-th delay unit with the i-th preset filter coefficient w at the current time k. i-1 (k) multiply and output w i-1 (k)*x(k-i+1); using the low-frequency compensation unit to output the preset low-frequency signal b(k) at the current moment k;

[0024] S3: using an adder to sum the output signals of each multiplier and the output signal of the low-frequency compensation unit and output y(k);

[0025] S4: using the decision device to generate a decision signal d(k) at the current time k based on the output signal y(k) of the adder at the current time k;

[0026] S5: Calculate the error signal e(k) at the current time k using the subtractor based on the formula e(k)=d(k)-y(k);

[0027] S6: Using the error signal e(k) at the current time k and the i-th preset filter coefficient w at the current time k i-1 (k) Update the preset filter coefficients w at the next time k+1 i-1 (k+1); using the error signal e(k) at the current moment k and the preset low-frequency signal b(k) at the current moment k, update the preset low-frequency signal b(k+1) at the next moment k+1.

[0028] In one embodiment, the error signal e(k) at the current time k and the i-th preset filter coefficient w at the current time k are used. i-1 (k) Update the preset filter coefficients w at the next time k+1 i-1 (k+1), including:

[0029] Using the formula w(k+1)=w(k)+μ w e(k)x * (k) Calculate the i-th preset filter coefficient w at the next moment k+1 i-1 (k+1), μ w is the update rate of the filter coefficients, x * (k) is the conjugate of x(k).

[0030] In one embodiment, the method of updating the preset low-frequency signal b(k+1) at the next moment k+1 by using the error signal e(k) at the current moment k and the preset low-frequency signal b(k) at the current moment k comprises:

[0031] Using the formula b(k+1)=b(k)+μ b e(k) calculates the preset low-frequency signal b(k+1) at the next moment k+1; where μ b is the update rate of the low frequency compensation unit.

[0032] In one embodiment, the output of the adder is:

[0033] y(k)=[w0(k),w1(k),...,w L-1 (k)] * ×[x(k),x(k-1),...,x(k-L+1)] T +b(k);

[0034] The superscript * stands for conjugate and T stands for transpose.

[0035] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0036] (1) The present invention provides a decision feedback adaptive equalizer with a low-frequency compensation unit. The equalizer makes up for the defect that the traditional decision feedback equalizer cannot process signal-independent noise by superimposing the signal b(k) output by the low-frequency compensation unit with the output signals of multiple multipliers, thereby realizing effective compensation for multipath interference signals. Furthermore, the output signal y(k) of the adder at the current moment k is used to generate the decision signal d(k) at the current moment k, and the error signal e(k) at the current moment k is calculated using the formula e(k)=d(k)-y(k) to update the filter coefficient and low-frequency signal compensation amount at the next moment, so as to adapt to the change of the channel; finally, the problem that the traditional decision feedback adaptive equalizer in the intensity modulation direct detection communication system cannot correctly equalize the multipath interference signal is solved, and the operation stability of the communication system is increased.

[0037] (2) This solution provides the use of the error signal e(k) at the current time k and the i-th preset filter coefficient w at the current time k. i-1 (k) Set the i-th preset filter coefficient w at the next moment k+1 i-1 (k+1); preferably, using the formula w(k+1)=w(k)+μ w e(k)x * (k) is calculated; associating the filter coefficient at the next moment with the current received signal and the update rate can better adapt to channel changes.

[0038] (3) This solution uses the error signal e(k) at the current moment k and the preset low-frequency signal b(k) at the current moment k to set the preset low-frequency signal b(k+1) at the next moment k+1; preferably, the formula b(k+1)=b(k)+μ b e(k) is calculated; considering the update rate of the low-frequency compensation signal at the next moment k+1 and the influence of the error signal at the current moment on the preset low-frequency signal b(k) at the current moment k, the operational stability of the communication system can be enhanced.

[0039] (4) In this scheme, the output of the adder is expressed as: y(k) = [w0(k), w1(k), ..., w L-1 (k)] * ×[x(k),x(k-1),...,x(k-L+1)] T +b(k); intuitively demonstrates that the output signal b(k) of the low-frequency compensation unit is superimposed on the output signals of the multiple multipliers to achieve compensation of multipath interference signals.

[0040] (5) This solution provides a decision feedback adaptive equalization method with a low-frequency compensation unit. While emphasizing the adaptive characteristics of the algorithm, it also pays special attention to its computational complexity and resource consumption. By introducing the low-frequency compensation unit, the overall complexity of the algorithm is controlled while improving the system's ability to suppress multipath interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a schematic diagram of the structure of a decision feedback adaptive equalizer with a low-frequency compensation unit provided in Embodiment 1 of the present invention;

[0042] Figure 2 is a flow chart of a decision feedback adaptive equalization method with a low-frequency compensation unit provided in Embodiment 2 of the present invention;

[0043] Figure 3a It is a schematic diagram of the effect of the received multipath interference signal and the low-frequency component tracked by the decision feedback adaptive equalization method provided by Embodiment 2 of the present invention;

[0044] Figure 3b It is a schematic diagram of a signal restored by the decision feedback adaptive equalization method provided by Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0046] Example 1

[0047] This embodiment provides a decision feedback adaptive equalizer with a low frequency compensation unit, such as Figure 1 As shown, it includes: multiple delay units, multiple multipliers, a low-frequency compensation unit, an adder, a judge, a subtractor and a controller.

[0048] A plurality of sequentially connected delay units, the first delay unit is used to delay the received signal x(k) of the decision feedback adaptive equalizer corresponding to the current time k and output x(k-1), the i-th delay unit is used to delay the output signal of the i-1-th delay unit and output x(k-i+1). i∈[1, L-1], L-1 is the total number of delay units.

[0049] A plurality of multipliers are connected to the plurality of delay units in a one-to-one correspondence, the input end of each multiplier is connected to the input end of the corresponding delay unit, and the i-th multiplier is used to multiply the input signal x(k-i+1) of the i-th delay unit with the i-th preset filter coefficient w at the current time k i-1 (k) multiply and output w i-1 (k)*x(k-i+1).

[0050] The low frequency compensation unit is used to output the preset low frequency signal b(k) at the current time k. Among them, low frequency is relative to the signal, W*X, where X is the input signal, and its frequency is very high. b(k) is signal-independent and processes the carrier beat frequency noise, so b(k) is called the output signal of the low frequency compensation unit.

[0051] The adder is connected to the output ends of the multiple multipliers and the output end of the low-frequency compensation unit, and is used to sum the output signals of the multiple multipliers and the output signal of the low-frequency compensation unit and output y(k).

[0052] The decider is connected to the output end of the adder and is used to generate a decision signal d(k) at the current moment k by using the output signal y(k) of the adder at the current moment k.

[0053] The subtractor is connected to the output end of the adder and the output end of the decision device, and is used to calculate the error signal e(k) at the current moment k using the formula e(k)=d(k)-y(k).

[0054] A controller connected to the subtractor, multiple multipliers and the low-frequency compensation unit is used to update each preset filter coefficient w at the next moment k+1 using the error signal e(k) at the current moment k i-1 (k+1) and a preset low frequency signal b(k+1).

[0055] In one embodiment, the i-th preset filter coefficient w at the next time k+1 i-1 (k+1) is the error signal e(k) at the current time k and the i-th preset filter coefficient w at the current time k i-1 (k) Set.

[0056] In one embodiment, the formula w(k+1)=w(k)+μ w e(k)x * (k) Calculate the i-th preset filter coefficient w at the next moment k+1 i-1 (k+1), μ w is the update rate of the filter coefficients, x * (k) is the conjugate of x(k).

[0057] In one embodiment, the preset low frequency signal b(k+1) of the low frequency compensation unit at the next time k+1 is set by using the error signal e(k) at the current time k and the preset low frequency signal b(k) at the current time k.

[0058] In one embodiment, the formula b(k+1)=b(k)+μ b e(k) calculates the preset low-frequency signal b(k+1) at the next moment k+1. b is the update rate of the low frequency compensation unit.

[0059] In one embodiment, the output of the adder is:

[0060] y(k)=[w0(k),w1(k),...,w L-1 (k)] * ×[x(k),x(k-1),...,x(k-L+1)] T +b(k).

[0061] The superscript * stands for conjugate and T stands for transpose.

[0062] Example 2

[0063] This embodiment provides a decision feedback adaptive equalization method with a low frequency compensation unit, such as Figure 2 As shown, it includes: steps S1-S6. It should be noted that,

[0064] S1: When the received signal x(k) is received at the current time k, it is delayed and output using multiple sequentially connected delay units. The i-th delay unit is used to delay the output signal of the i-1-th delay unit and output x(k-i+1). i∈[1, L-1], L-1 is the total number of delay units.

[0065] S2: Use each multiplier to multiply the input signal of the corresponding delay unit with the corresponding preset filter coefficient and output it. The i-th multiplier multiplies the input signal x(k-i+1) of the i-th delay unit with the i-th preset filter coefficient w at the current time k. i-1 (k) multiply and output w i-1 (k)*x(k-i+1). The low-frequency compensation unit is used to output the preset low-frequency signal b(k) at the current moment k.

[0066] S3: using an adder to sum the output signals of each multiplier and the output signal of the low-frequency compensation unit and output y(k).

[0067] S4: Generate a decision signal d(k) at the current time k using a decision device based on the output signal y(k) of the adder at the current time k.

[0068] S5: Calculate the error signal e(k) at the current time k using the subtractor based on the formula e(k)=d(k)-y(k).

[0069] S6: Using the error signal e(k) at the current time k and the i-th preset filter coefficient w at the current time k i-1 (k) Update the preset filter coefficients w at the next time k+1 i-1 The preset low-frequency signal b(k+1) at the next moment k+1 is updated by using the error signal e(k) at the current moment k and the preset low-frequency signal b(k) at the current moment k.

[0070] It should be noted that the decision feedback adaptive equalization method with a low-frequency compensation unit provided in this embodiment has a good restoration effect on multipath interference signals. Figure 3a and Figure 3b shown. Figure 3a The received multipath interference signal and the low-frequency components tracked by the decision feedback adaptive equalization method are shown; Figure 3b Displays the signal after recovery using the decision feedback adaptive equalization method.

[0071] While emphasizing the adaptive characteristics of the algorithm, this embodiment also pays special attention to its computational complexity and resource consumption. By introducing a low-frequency compensation unit, the overall complexity of the algorithm is controlled while improving the system's ability to suppress multipath interference. By effectively optimizing resource utilization, rapid convergence in a complex channel environment is achieved, ensuring a balance between compensation effect and computational efficiency, thereby meeting the requirements of high efficiency and low resource occupancy.

[0072] In one embodiment, the formula w(k+1)=w(k)+μ w e(k)x * (k) Calculate the i-th preset filter coefficient w at the next moment k+1 i-1 (k+1), μ w is the update rate of the filter coefficients, x * (k) is the conjugate of x(k).

[0073] In one embodiment, the formula b(k+1)=b(k)+μ b e(k) calculates the preset low-frequency signal b(k+1) at the next moment k+1. b is the update rate of the low frequency compensation unit.

[0074] In one embodiment, the output of the adder is:

[0075] y(k)=[w0(k),w1(k),...,w L-1 (k)]* ×[x(k),x(k-1),...,x(k-L+1)] T +b(k).

[0076] The superscript * stands for conjugate and T stands for transpose.

[0077] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A decision feedback adaptive equalizer with a low frequency compensation unit, characterized in that: include: A plurality of sequentially connected delay units, wherein the first delay unit is used to delay the received signal x(k) of the decision feedback adaptive equalizer corresponding to the current time k and output x(k-1), and the i-th delay unit is used to delay the output signal of the i-1-th delay unit and output x(k-i+1); i∈[1, L-1], L-1 is the total number of the delay units; A plurality of multipliers are connected to the plurality of delay units in a one-to-one correspondence, the input end of each multiplier is connected to the input end of the corresponding delay unit, and the i-th multiplier is used to multiply the input signal x(k-i+1) of the i-th delay unit by the i-th preset filter coefficient w at the current time k i-1 (k) multiply and output w i-1 (k)*x(k-i+1); A low-frequency compensation unit, used to output a preset low-frequency signal b(k) at the current moment k; an adder connected to the output ends of the plurality of multipliers and the output end of the low-frequency compensation unit, and configured to sum the output signals of the plurality of multipliers and the output signal b(k) of the low-frequency compensation unit and output y(k); A decision device, connected to the output end of the adder, for generating a decision signal d(k) at the current time k using the output signal y(k) of the adder at the current time k; A subtractor, connected to the output end of the adder and the output end of the decision device, for calculating the error signal e(k) at the current moment k using the formula e(k)=d(k)-y(k); A controller connected to the subtractor, the plurality of multipliers and the low-frequency compensation unit, configured to update each preset filter coefficient w at the next moment k+1 using the error signal e(k) at the current moment k i-1 (k+1) and a preset low frequency signal b(k+1).

2. The decision feedback adaptive equalizer with a low frequency compensation unit according to claim 1, characterized in that: The i-th preset filter coefficient w at the next moment k+1 i-1 (k+1) is the error signal e(k) at the current time k and the i-th preset filter coefficient w at the current time k i-1 (k) Set.

3. The decision feedback adaptive equalizer with a low frequency compensation unit as claimed in claim 2, characterized in that: Using the formula w(k+1)=w(k)+μ w e(k)x * (k) Calculate the i-th preset filter coefficient w at the next moment k+1 i-1 (k+1), μ w is the update rate of the filter coefficients, x * (k) is the conjugate of x(k).

4. The decision feedback adaptive equalizer with a low frequency compensation unit according to claim 1, characterized in that: The preset low frequency signal b(k+1) of the low frequency compensation unit at the next time k+1 is set by using the error signal e(k) at the current time k and the preset low frequency signal b(k) at the current time k.

5. The decision feedback adaptive equalizer with a low frequency compensation unit as claimed in claim 4, characterized in that: Using the formula b(k+1)=b(k)+μ b e(k) calculates the preset low-frequency signal b(k+1) at the next moment k+1; where μ b is the update rate of the low frequency compensation unit.

6. The decision feedback adaptive equalizer with a low frequency compensation unit according to claim 1, characterized in that: The output of the adder is: y(k)=[w0(k),w1(k),...,w L-1 (k)] * ×[x(k),x(k-1),...,x(k-L+1)] T +b(k); The superscript * stands for conjugate and T stands for transpose.

7. A decision feedback adaptive equalization method with a low frequency compensation unit, characterized in that: The decision feedback adaptive equalizer with a low frequency compensation unit as claimed in any one of claims 1 to 6 comprises: S1: When the received signal x(k) is received at the current time k, a plurality of sequentially connected delay units are used to delay and output; the i-th delay unit is used to delay the output signal of the i-1-th delay unit and output x(k-i+1); i∈[1, L-1], L-1 is the total number of the delay units; S2: Use each multiplier to multiply the input signal of the corresponding delay unit with the corresponding preset filter coefficient and output it. The i-th multiplier multiplies the input signal x(k-i+1) of the i-th delay unit with the i-th preset filter coefficient w at the current time k. i-1 (k) multiply and output w i-1 (k)*x(k-i+1); using the low-frequency compensation unit to output the preset low-frequency signal b(k) at the current moment k; S3: using an adder to sum the output signals of each multiplier and the output signal of the low-frequency compensation unit and output y(k); S4: using the decision device to generate a decision signal d(k) at the current time k based on the output signal y(k) of the adder at the current time k; S5: Calculate the error signal e(k) at the current time k using the subtractor based on the formula e(k)=d(k)-y(k); S6: Using the error signal e(k) at the current time k and the i-th preset filter coefficient w at the current time k i-1 (k) Update the preset filter coefficients w at the next time k+1 i-1 (k+1); using the error signal e(k) at the current moment k and the preset low-frequency signal b(k) at the current moment k, update the preset low-frequency signal b(k+1) at the next moment k+1.

8. The decision feedback adaptive equalization method with a low frequency compensation unit according to claim 7, characterized in that: The error signal e(k) at the current time k and the i-th preset filter coefficient w at the current time k are used. i-1 (k) Update the preset filter coefficients w at the next time k+1 i-1 (k+1), including: Using the formula w(k+1)=w(k)+μ w e(k)x * (k) Calculate the i-th preset filter coefficient w at the next moment k+1 i-1 (k+1), μ w is the update rate of the filter coefficients, x * (k) is the conjugate of x(k).

9. The decision feedback adaptive equalization method with a low frequency compensation unit according to claim 7, characterized in that: The method of updating the preset low-frequency signal b(k+1) at the next moment k+1 by using the error signal e(k) at the current moment k and the preset low-frequency signal b(k) at the current moment k comprises: Using the formula b(k+1)=b(k)+μ b e(k) calculates the preset low-frequency signal b(k+1) at the next moment k+1; where μ b is the update rate of the low frequency compensation unit.

10. The decision feedback adaptive equalization method with a low frequency compensation unit according to claim 7, characterized in that: The output of the adder is: y(k)=[w0(k),w1(k),...,w L-1 (k)] * ×[x(k),x(k-1),...,x(k-L+1)] T +b(k); The superscript * stands for conjugate and T stands for transpose.

Images