SerDes sending end equalization optimization method based on simulated annealing algorithm

Abstract

The invention discloses a SerDes sending end equalization optimization method based on a simulated annealing algorithm, and belongs to the field of digital signal processing, and the method comprises the steps: initializing the algorithm, including configuring the number of iterations and an initial solution; automatically setting other parameters of the algorithm, including an initial temperature and an attenuation coefficient; new solutions are generated as simulation parameters, the advantages and disadvantages of the new solutions are judged, and the new solutions need to meet constraint conditions; simulating a SerDes system, taking a logarithm of which the bit error rate takes 10 as the bottom as an energy index, and judging whether the change value of energy is less than 0 or not; whether an inferior solution is accepted or not is judged, that is, when the energy change value is larger than 0, the current simulation solution is still likely to be accepted, and the capability of jumping out of a local optimal solution is achieved; judging whether to terminate, and when a termination condition is met, terminating the iteration process; and outputting an optimal solution. According to the method, automatic configuration based on the simulated annealing algorithm is realized, and the SerDes design efficiency is greatly improved.

Description

Technical Field

[0001] The invention belongs to the field of digital signal processing, and in particular relates to a SerDes (Serializer and Deserializer) transmitter equalization optimization method based on a simulated annealing algorithm. Background Art

[0002] In the current era of the Internet of Everything, high-speed serial systems have become the mainstream of data transmission and exchange, and have been widely used in various industries. Serializer and Deserializer (SerDes) is an important technology used in high-speed serial systems. It is mainly used to convert data from parallel transmission to serial transmission (serialization) and restore serial transmission data to parallel data (deserialization). This technology is widely used in various communication systems, data centers, computer systems and embedded devices, especially in situations where high bandwidth and long-distance transmission are required.

[0003] When the data transmission rate is high, such as at the Gbit / s level, due to skin effect, dielectric loss and other reasons, the signal will be severely attenuated when it reaches the receiving end through the transmission line, resulting in the receiving end being unable to receive the correct signal. Usually, equalization technology is used at the transmitting and receiving ends of SerDes to solve this problem. Equalization at the transmitting end is also called pre-emphasis or de-emphasis. It is a feed forward equalizer (FFE), which is essentially a finite impulse response (FIR) filter. Equalization at the receiving end generally uses an adaptive algorithm for optimization, while there is no corresponding adaptive algorithm for equalization at the transmitting end. Equalization at the transmitting end can only be performed by manually adjusting the equalization parameters.

[0004] for Figure 2 The FFE shown is an FIR filter with 5 taps. If the main tap is fixed, there are 4 taps that need to be manually configured, and they need to be normalized after adjustment. To find the optimal tap configuration, you need to manually adjust the size of the 4 taps and observe the test results. Each test takes a certain amount of time, so it will consume a lot of man-hours. In the design and product prototype stage, the transmitter equalization is mostly manually preset, with a low degree of automation, which is very different from the receiver equalization. Therefore, in the context of the current manual preset of the transmitter equalization, in order to improve the design efficiency, there is an urgent need for an automatic optimization method to optimize the transmitter equalization. Summary of the invention

[0005] To solve the above technical problems, the present invention provides a SerDes transmitter equalization optimization method based on a simulated annealing algorithm, which takes the logarithm of the bit error rate of the SerDes system with base 10 as the system energy index of the algorithm, and automatically searches for the transmitter equalization parameters that make the system bit error rate as low as possible, that is, the energy as low as possible. Compared with manual adjustment, the method has a high degree of automation and higher robustness.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A SerDes transmitter equalization optimization method based on a simulated annealing algorithm comprises the following steps:

[0008] Step 1: Configure the number of iterations of the simulated annealing algorithm and set an all-zero or random initial solution, which represents a set of tap coefficients of the FFE at the sender.

[0009] Step 2: Automatically set other parameters of the simulated annealing algorithm, including the initial temperature, based on the configured parameters and attenuation coefficient ;

[0010] Step 3: Generate a new solution within the constraints. If the solution of the previous simulation is not accepted, a new solution is generated with the old solution as the starting point. If the solution of the previous simulation is accepted, a new solution is generated with the accepted solution as the starting point.

[0011] Step 4: Simulate the new solution in step 3 and determine the energy change value of the system Is it less than 0? If the result is "yes", accept the solution of the current simulation and go to step 6. If the result is "no", go to step 5.

[0012] Step 5: Judgement Is it true, T: the current temperature T of the simulated annealing algorithm; rand: a random number uniformly distributed between 0 and 1; that is, The probability of accepting the solution of the current simulation is , even if it has a worse simulation result than the old solution. If the judgment result is "yes", the solution of the current simulation is accepted and the process goes to step 6. If the simulation result is "no", the solution of the current simulation is rejected and the process goes to step 6.

[0013] Step 6: Determine whether the termination condition is met, that is, whether the specified number of iterations has been reached; if the result is "yes", output the optimal solution; if the result is "no", multiply the current temperature T by the attenuation coefficient And go to step 3;

[0014] Step 7: Output the optimal solution and use the currently selected optimal solution as the tap coefficient of the FFE at the transmitting end.

[0015] In step 3, the specific implementation is as follows:

[0016] Step 31, determine whether the solution of the last iterative simulation is accepted;

[0017] Step 32: If the result of step 31 is "yes", taking the solution of the last simulation as the starting point, for each tap value of the FFEtap coefficient of the transmitting end, add a random value as the tap value of the new solution;

[0018] Step 33: If the result of step 31 is "no", taking the old solution as the starting point, for each tap value of the transmitting end FFE tap coefficient, add a random value as the tap value of the new solution;

[0019] Step 34: for the generated new solution, determine whether it satisfies the set constraints. If so, proceed to step 4; otherwise, proceed to step 32 or step 33 according to the judgment result of step 31 to regenerate a new solution until the constraints are met.

[0020] In step 4, the specific implementation is as follows:

[0021] Step 41: Use the generated new solution, i.e., the transmitting end FFE tap coefficient, as a configuration parameter of the SerDes simulation system, and perform bit-by-bit simulation on the SerDes;

[0022] Step 42: The bit error rate (BER) can be calculated from the simulation results, and its logarithm with base 10 is taken as the energy value to determine the energy change value. Is it less than 0?

[0023] Step 43: If the energy change value If it is less than 0, it means that the bit error rate of the current simulation result is lower, then the solution of the current simulation is accepted, otherwise go to step 5.

[0024] On the other hand, the present invention provides a SerDes transmitter equalization optimization system based on a simulated annealing algorithm, comprising:

[0025] Configuration unit, used to configure the number of iterations and initial solution of the simulated annealing algorithm, the initial solution represents the tap coefficient of the FFE at the sending end; set the initial temperature of the simulated annealing algorithm and attenuation coefficient ;

[0026] A generation unit, used for generating new solutions based on the constraints of the simulated annealing algorithm;

[0027] The judgment unit is used to use the new solution to simulate the system and make the first judgment to determine the energy change value of the system. Is it less than 0? If the first judgment result is "yes", the current simulation solution is accepted and judged whether it meets the specified number of iterations; if the first judgment result is "no", the second judgment is performed to judge whether Is it established, where T represents the current temperature of the simulated annealing algorithm, and rand represents a random number uniformly distributed between 0 and 1. If the second judgment result is "yes", the current simulation solution is accepted and it is judged whether it meets the specified number of iterations. If the second judgment result is "no", the current simulation solution is rejected and it is judged whether it meets the specified number of iterations. If it meets the specified number of iterations, the optimal solution is output. If it does not meet the specified number of iterations, it returns to the generation unit.

[0028] The output unit uses the currently selected optimal solution as the tap coefficient of the FFE at the transmitting end.

[0029] In a third aspect, the present invention provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the aforementioned SerDes transmitter equalization optimization method based on a simulated annealing algorithm.

[0030] In a fourth aspect, the present invention provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, enables the processor to implement the aforementioned SerDes transmitter equalization optimization method based on a simulated annealing algorithm.

[0031] The beneficial effects of the present invention are:

[0032] (1) The present invention can automatically adjust the tap coefficient of the FFE at the transmitting end, with a high degree of automation, saving manpower and material resources.

[0033] (2) The present invention can realize automatic setting of the parameters of the simulated annealing algorithm, including the initial temperature and attenuation coefficient , so that the probability of accepting an inferior solution varies with the number of iterations within a reasonable range. The purpose is to ensure that every iteration is fully utilized and the global optimal solution is found as much as possible.

[0034] (3) Compared with other parameter optimization algorithms such as the minimum gradient descent algorithm, the algorithm used in the present invention can prevent falling into the local optimal solution and has a greater possibility of finding the global optimal solution because it has the ability to accept inferior solutions.

[0035] (4) Certain constraints are set for the search for solutions. According to the inherent characteristics of channel attenuation and PCIE specifications, the FFE tap coefficient of the SerDes transmitter is constrained, which can effectively improve the efficiency of the algorithm and greatly reduce the number of iterations to find the optimal solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A flow chart of a SerDes transmitter equalization optimization method based on a simulated annealing algorithm according to the present invention;

[0037] Figure 2 Schematic diagram of FFE implementation with two forward taps and two backward taps;

[0038] Figure 3 The relationship between the change process of the probability of accepting an inferior solution and the initial temperature;

[0039] Figure 4 The relationship between the changing process of the probability of accepting an inferior solution and the attenuation coefficient;

[0040] Figure 5 This is a schematic diagram of the SerDes simulation system;

[0041] Figure 6 This is a graph showing the relationship between the bit error rate and tap coefficient and the number of iterations during the simulation process. DETAILED DESCRIPTION

[0042] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0043] like Figure 1 As shown in the figure, the process of the SerDes transmitter equalization optimization method based on the simulated annealing algorithm of the present invention mainly configures the number of iterations and the initial solution, automatically sets the algorithm parameters, generates a new solution within the constraint conditions, simulates and determines the energy change value Is it less than 0? Whether it is greater than rand, T is the temperature, rand is a random number between 0 and 1, whether the termination condition is met, and the logic of outputting the optimal solution is executed. The specific process of each part is explained below.

[0044] Step 1: Configure the number of iterations and initial solution: Configure the number of iterations of the simulated annealing algorithm and set an all-zero or random initial solution, which represents a set of tap coefficients of the FFE at the sender. Figure 2The figure shows the implementation principle diagram of FFE with two pre-tap and two post-tap. Since the driving capability of the transmitter is limited, it is necessary to ensure that the sum of the absolute values ​​of the FFE coefficients at the transmitter does not exceed a specific value. Therefore, the main tap c(0) can be fixed, the other four tap values ​​can be adjusted, and finally normalized. The number of iterations can be appropriately increased. The larger the number of iterations, the greater the possibility of finding a better solution.

[0045] Step 2: Automatically set algorithm parameters: Automatically set other parameters of the simulated annealing algorithm, including the initial temperature, according to the configured parameters and attenuation coefficient According to the characteristics of the simulated annealing algorithm, as the iteration process proceeds, the probability of accepting an inferior solution will gradually decrease, and this decrease process is affected by the initial temperature. , attenuation coefficient impact. Figure 3 and Figure 4 The change process of the probability of accepting inferior solutions and the initial temperature are shown respectively , attenuation coefficient The temperature attenuation refers to the temperature multiplied by the attenuation coefficient after each iteration as the temperature for the next iteration. It can be seen from the figure that these two parameters have a great influence on the probability of accepting inferior solutions in the iteration process. Whenever the total number of iterations changes, in order to make the probability of accepting inferior solutions change reasonably throughout the iteration process, the initial temperature and attenuation coefficient need to be re-specified.

[0046] The following is a method for automatically setting parameters. Assume that the total number of iterations is N, and stipulate that when the number of iterations n = 0.2N, the probability of accepting an inferior solution is P = 0.95; when the number of iterations n = 0.8N, the probability of accepting an inferior solution is P = 0.05. Then the simultaneous equations are as follows (1):

[0047] (1)

[0048] in It is the change value of energy, similar to the accuracy of the algorithm, indicating the sensitivity to the change of BER magnitude, and is generally set to 0.1.

[0049] According to formula (1), we can get:

[0050] (2)

[0051] According to formula (2), the initial temperature of the simulated annealing algorithm , attenuation coefficient Perform automatic settings.

[0052] Step 3, Generate a new solution within the constraints: Generate a new solution within the constraints. If it is the first iteration currently, then use the initial solution as the starting point to generate a new solution for subsequent simulations; if the current iteration is the second iteration or later, then generate a new solution based on the acceptance situation of the solution in the previous iteration: If the solution of the previous iteration simulation is not accepted, then use the optimal solution before the previous iteration started as the old solution, and use this as the starting point to generate a new solution. For example, for each tap value of the transmitter FFE tap coefficients, add a random value as the tap value of the new solution; if the solution of the previous simulation is accepted, then use the accepted solution as the starting point to generate a new solution. For example, for each tap value of the transmitter FFE tap coefficients, add a random value as the tap value of the new solution. Referring to the PCIE specification, taking the transmitter FFE with 5 taps as an example (without loss of generality, it can be 2N + 1, and here only 5 are taken as an example), the tap coefficients c(-2), c(-1), c(0), c(1), c(2) first need to satisfy:

[0053] (3)

[0054] Secondly, define the maximum amplitude voltage As shown in Equation (4), define the de-emphasis voltage As shown in Equation (5), the relationship between the two voltages needs to satisfy Equation (6), where represents 、 the magnitude relationship between, measured in decibels.

[0055] (4)

[0056] (5)

[0057] (6)

[0058] Finally, is generally small. Therefore, if the magnitude of c(0) is 64, then Equation (7) needs to be satisfied:

[0059] (7)

[0060] The method for generating a new solution is: fix c(0), and add a random increment to the other coefficients except c(0) on the original basis. It can be considered that both the new solution and the old solution are points in a multi-dimensional space. Generating a new solution is to let the old solution "take a step" in the multi-dimensional space, where The maximum step length of should be set smaller than because the absolute value of .

[0061] Generating a new solution under the specified constraints can greatly improve the efficiency of the algorithm, avoid generating an unreasonable solution, and enable the algorithm to quickly find the optimal solution. If the new solution does not meet the constraints, a new solution needs to be generated again until the constraints are met.

[0062] Step 4: Simulate and judge Is it less than 0: The new solution in simulation step 3 and the value of the energy change of the system Is it less than 0? If the result is "yes", accept the current simulation solution and go to step 6. If the result is "no", go to step 5. Use the generated new solution as the tap coefficient of the transmitting end FFE to simulate the SerDes system bit by bit. Figure 5 As shown in the figure, SerDes TX is the transmitter, including the FFE of the transmitter, Channel is the channel modeled using s parameters, and SerDes RX is the receiver, including DFE (decision feedback equalization), FFE and other adaptive equalizers. Assume that the current optimal bit error rate is BER, and the bit error rate obtained in this simulation is BER', then The calculation of is given by formula (8):

[0063] (8)

[0064] Calculated Afterwards, if it is less than 0, it means that the current simulation solution has a lower bit error rate, and the current simulation solution is accepted; otherwise, if it is not less than 0, it means that the current simulation solution has a higher bit error rate, and then it is necessary to further determine whether to accept the current simulation solution.

[0065] Step 5: Judgement Is it greater than rand? If the judgment result is "yes", accept the solution of the current simulation and go to step 6. If the simulation result is "no", reject the solution of the current simulation and go to step 6. T represents the current temperature of the simulated annealing algorithm. First, you need to generate a random number rand uniformly distributed between 0 and 1, and calculate Is the value of greater than rand? exist When T is not less than 0 and T is greater than 0, it is a positive number less than 1, so the probability that rand is less than it is So the final result is The probability of accepting the inferior solution is The probability of accepting the solution of the current simulation even if it has worse simulation results than the old solution.

[0066] Step 6: Determine whether the termination condition is met. According to the set termination condition, you can choose whether the maximum number of convergences is reached, whether the specified bit error rate is reached, etc. as the termination condition. If the termination condition is not met, the temperature attenuation operation needs to be performed, as shown in formula (9), where T is the current temperature, is the temperature after attenuation. If the termination condition is reached, the current optimal solution is output. If the result is "no", the current temperature T is multiplied by the attenuation coefficient And go to step 3;

[0067] (9)

[0068] Step 7: Output the optimal solution. Feedback the optimal solution to the FFE at the SerDes transmitter as a set of optimal tap coefficients for the FFE at the transmitter.

[0069] Example

[0070] The SerDes simulation system generates a set of PRBS31 signals to simulate PAM4 signals. The signals are sent out after being pre-equalized by FFE at the transmitting end. After channel attenuation, the signals are received at the receiving end and equalized at the receiving end, including CTLE (continuous time linear equalization), DFE, FFE, etc. Finally, the bit errors are counted to calculate the bit error rate.

[0071] The number of iterations is set to 50, and the transmitter FFE uses one forward tap and one backward tap. Not used, always 0. Initialize the tap coefficient (unnormalized):

[0072] ;

[0073] After initialization, the algorithm automatically configures the remaining parameters:

[0074] ;

[0075] When generating a new solution, the main tap c(0) is fixed to 64. A random integer between -3 and 3 is added, c(-1) must satisfy formula (3), and according to formula (6) we can deduce The following conditions must also be met:

[0076] ;

[0077] Finally, after 50 iterations, we get The final result:

[0078] ;

[0079] from Figure 6It can be seen that with the increase in the number of iterations, the bit error rate of the system becomes lower and lower and tends to a constant value, and the tap coefficient also converges to a constant value, indicating the correctness of the SerDes transmitter equalization optimization method based on the simulated annealing algorithm of the present invention.

[0080] On the other hand, the present invention provides a SerDes transmitter equalization optimization system based on a simulated annealing algorithm, wherein each module included in the system can implement each step of the aforementioned method, specifically including:

[0081] Configuration unit, used to configure the number of iterations and initial solution of the simulated annealing algorithm, the initial solution represents the tap coefficient of the FFE at the sending end; set the initial temperature of the simulated annealing algorithm and attenuation coefficient ;

[0082] A generation unit, used for generating new solutions based on the constraints of the simulated annealing algorithm;

[0083] The judgment unit is used to use the new solution to simulate the system and make the first judgment to determine the energy change value of the system. Is it less than 0? If the first judgment result is "yes", the current simulation solution is accepted and judged whether it meets the specified number of iterations; if the first judgment result is "no", the second judgment is performed to judge whether Is it true, where T represents the current temperature of the simulated annealing algorithm, and rand represents a random number uniformly distributed between 0 and 1. If the second judgment result is "yes", the current simulation solution is accepted and it is judged whether it meets the specified number of iterations. If the second judgment result is "no", the current simulation solution is rejected and it is judged whether it meets the specified number of iterations. If it meets the specified number of iterations, the optimal solution is output. If it does not meet the specified number of iterations, it returns to the generation unit.

[0084] The output unit uses the currently selected optimal solution as the tap coefficient of the FFE at the transmitting end.

[0085] In a third aspect, the present invention provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the aforementioned SerDes transmitter equalization optimization method based on simulated annealing algorithm.

[0086] In a fourth aspect, the present invention provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, enables the processor to implement the aforementioned SerDes transmitter equalization optimization method based on a simulated annealing algorithm.

[0087] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A SerDes transmitter equalization optimization method based on simulated annealing algorithm, characterized in that: The method comprises: Step 1: Configure the number of iterations and the initial solution of the simulated annealing algorithm, where the initial solution represents the tap coefficient of the FFE at the transmitting end; Step 2: Set the initial temperature of the simulated annealing algorithm and attenuation coefficient α; Step 3: Generate a new solution based on the constraints of the simulated annealing algorithm; Step 4: Use the new solution to simulate the system and determine the energy change value of the system Is it less than 0? If the result is "yes", accept the current simulation solution and go to step 6; if the result is "no", go to step 5; Step 5: Judgement Is it true, where T represents the current temperature of the simulated annealing algorithm, and rand represents a random number uniformly distributed between 0 and 1; if the judgment result is "yes", the current simulation solution is accepted and the process goes to step 6; if the simulation result is "no", the current simulation solution is rejected and the process goes to step 6; Step 6: Determine whether the specified number of iterations is met. If the result is "yes", output the optimal solution. If the result is "no", go to step 3. Step 7: Use the currently selected optimal solution as the tap coefficient of the FFE at the transmitting end.

2. The SerDes transmitter equalization optimization method based on simulated annealing algorithm according to claim 1, characterized in that: In step 2, the initial temperature and attenuation coefficient of the simulated annealing algorithm are automatically set according to the following formula: ； Where N is the number of iterations of the configuration, is the change in energy.

3. The SerDes transmitter equalization optimization method based on simulated annealing algorithm according to claim 1, characterized in that: The step 3 comprises: Step 31, determine whether the solution of the last iterative simulation is accepted; Step 32: If the result of step 31 is "yes", taking the solution of the last simulation as the starting point, for each tap value of the FFE tap coefficient at the transmitting end, add a random value as the tap value of the new solution; Step 33: If the result of step 31 is "no", taking the old solution as the starting point, for each tap value of the transmitting end FFE tap coefficient, add a random value as the tap value of the new solution; Step 34: for the generated new solution, determine whether it satisfies the set constraints. If so, proceed to step 4; otherwise, proceed to step 32 or step 33 according to the judgment result of step 31 to regenerate a new solution until the constraints are met.

4. The SerDes transmitter equalization optimization method based on simulated annealing algorithm according to claim 3, characterized in that: In step 34, assuming that the transmitting end FFE includes 2N+1 taps, the tap coefficients are c(-N), c(-(N-1)), ..., c(0), c(N-1), c(N), respectively, and the constraint condition is: （3） （6） （7） In the formula, Indicates the maximum amplitude voltage, represents the de-emphasis voltage, express , The relationship between the sizes is measured in decibels.

5. The SerDes transmitter equalization optimization method based on simulated annealing algorithm according to claim 1, characterized in that: The step 4 comprises: Step 41: Use the generated new solution as the configuration parameter of the SerDes simulation system to simulate the SerDes bit by bit; Step 42: Calculate the bit error rate based on the simulation results, take its logarithm with base 10 as the energy value, and determine the energy change value. Is it less than 0? Step 43: If the energy change value If it is less than 0, accept the solution of the current simulation; otherwise, go to step 5.

6. The SerDes transmitter equalization optimization method based on simulated annealing algorithm according to claim 5, characterized in that: The step 42 includes assuming that the current optimal bit error rate is BER, the bit error rate obtained by this simulation is BER', The calculation of is given by formula (8): （8）。 7. The SerDes transmitter equalization optimization method based on simulated annealing algorithm according to claim 1, characterized in that: In step 6, if the result is "no", the current temperature T is multiplied by the attenuation coefficient α and the process goes to step 3.

8. A SerDes transmitter equalization optimization device based on simulated annealing algorithm, characterized in that: include: Configuration unit, used to configure the number of iterations and initial solution of the simulated annealing algorithm, the initial solution represents the tap coefficient of the FFE at the sending end; set the initial temperature of the simulated annealing algorithm and attenuation coefficient α; A generation unit, used for generating new solutions based on the constraints of the simulated annealing algorithm; The judgment unit is used to use the new solution to simulate the system and make the first judgment to determine the energy change value of the system. Is it less than 0? If the first judgment result is "yes", the current simulation solution is accepted and it is judged whether it meets the specified number of iterations; if the first judgment result is "no", a second judgment is performed to judge whether Is it established, where T represents the current temperature of the simulated annealing algorithm, and rand represents a random number uniformly distributed between 0 and 1. If the second judgment result is "yes", the current simulation solution is accepted and it is judged whether it meets the specified number of iterations. If the second judgment result is "no", the current simulation solution is rejected and it is judged whether it meets the specified number of iterations. If it meets the specified number of iterations, the optimal solution is output. If it does not meet the specified number of iterations, it returns to the generation unit. The output unit uses the currently selected optimal solution as the tap coefficient of the FFE at the transmitting end.

9. An electronic device, characterized in that: include: one or more processors; A memory for storing one or more programs; Wherein, when one or more programs are executed by the one or more processors, the one or more processors implement a SerDes transmitter equalization optimization method based on a simulated annealing algorithm as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: Executable instructions are stored thereon, and when the instructions are executed by the processor, the processor can implement the SerDes transmitting end equalization optimization method based on the simulated annealing algorithm as described in any one of claims 1-7.

Images