Grid-based printed circuit board wiring design device and method and computer readable medium

By identifying and adjusting the contingency value of crowded grids in printed circuit board wiring design and redefining the optimal path, the grid congestion problem caused by multiple network wiring is solved, and automated wiring design is realized and cost and wiring length is reduced.

CN119067056APending Publication Date: 2024-12-03BOSCH AUTOMOTIVE PRODUCTS (SUZHOU) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310646403.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-06-02
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Prior Art In printed circuit board wiring design, wiring paths of multiple networks are prone to congestion in grids, resulting in wiring failure, and manual detours increase costs and wiring length.

Method used

By determining the initial generation value and occupation status of each grid by combining the path determination unit, the grid occupation state determination unit, the crowded grid determination unit and the generation value adjustment unit, identifying the crowded grid and increasing its generation value, re- Determine the optimal wiring path to avoid mesh congestion.

Benefits of technology

This enables automated wiring designs for multiple networks without significantly increasing the overall wiring length, avoiding grid congestion, ensuring successful wiring and reducing costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067056A_ABST
    Figure CN119067056A_ABST
Patent Text Reader

Abstract

A mesh-based printed circuit board routing design apparatus is proposed, comprising: a path determination unit for a) determining an optimal routing path for each of a plurality of networks based on an initial cost value of each mesh, each network comprising a pair of electronic devices to be connected by the routing path, the optimal wiring path represents a wiring path of which the sum of the cost values of the passing grids is minimum; a grid occupancy state determination unit for b) determining an occupancy state value of each grid occupied by the determined optimal routing path; a congested grid determination unit for c) determining a congested grid, the occupancy state value of which exceeds a state threshold; and a cost value adjustment unit for d) increasing the cost value of the crowded grid for at least some of the plurality of networks, where the path determination unit is further for e) re-determining the optimal routing path for the at least some of the plurality of networks based on the increased cost value. By increasing the cost value of the congested grid, the routing paths of some networks may be forced to bypass the congested grid, thereby enabling routing designs for multiple networks to be automatically implemented without causing grid congestion, while not significantly increasing the overall routing length.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a printed circuit board wiring design, and more particularly to a grid-based wiring design for a plurality of networks on a printed circuit board. Background Art

[0002] Manual wiring design of printed circuit boards is a tedious and time-consuming task. Automatic wiring design is a technology that can replace manual wiring design, which can realize automatic wiring based on some predetermined criteria. Considering the anti-electromagnetic interference performance and manufacturing cost, one of the criteria is that the wiring path between electronic devices should be as short as possible. The shortest path can be found for a pair of electronic devices to be connected by wiring based on the grid. In this technology, the printed circuit board is divided into a grid array, and each grid is assigned the same cost value, such as 1. For a pair of electronic devices, there may be multiple paths of varying lengths between them. The sum of the cost values ​​of all the grids passed by each path can be determined. For example, if a path passes through 5 grids, its sum of cost values ​​is determined to be 5. The path with the smallest sum of cost values ​​is the path that passes through the least grids, that is, the shortest path, and can therefore be determined as the best path between the pair of electronic devices.

[0003] Generally, the number of electronic devices on a printed circuit board is numerous and complex. Therefore, when designing the wiring, it is impossible to consider only one network including a pair of electronic devices. Automatic wiring design should be able to design wiring for multiple networks at the same time. After finding the shortest path for each network based on the grid, multiple shortest paths may pass through a certain grid (for example, a grid close to the center area of ​​the printed circuit board), which leads to grid congestion, that is, the number of wiring accommodated by the grid exceeds its maximum allowable capacity, resulting in wiring failure. In this case, it is usually necessary to rely on manual detours for some of the networks, which not only increases labor costs, but also may greatly increase the overall wiring length after detours.

[0004] Therefore, it remains desirable to provide improved printed circuit board routing designs. Summary of the invention

[0005] The present invention provides an improved printed circuit board routing design, which can automatically implement routing design for multiple nets without causing grid congestion and without significantly increasing the overall routing length.

[0006] According to one aspect of the present invention, a grid-based printed circuit board wiring design device is provided, comprising: a path determination unit, which is used to a) determine an optimal wiring path for each network in a plurality of networks based on an initial cost value of each grid, each network including a pair of electronic devices to be connected through a wiring path, and the optimal wiring path represents a wiring path that passes through a grid with a minimum sum of cost values; a grid occupancy state determination unit, which is used to b) determine an occupancy state value of each grid, the occupancy state value representing an occupancy state of a grid occupied by the determined optimal wiring path; a crowded grid determination unit, which is used to c) determine a crowded grid, the occupancy state value of the crowded grid exceeding a state threshold; and a cost value adjustment unit, which is used to d) increase the cost value of the crowded grid for at least some of the plurality of networks, wherein the path determination unit is further used to e) redetermine the optimal wiring path for the at least some of the plurality of networks based on the increased cost value.

[0007] According to another aspect of the present invention, a grid-based printed circuit board wiring design method is proposed, comprising: a) determining an optimal wiring path for each of a plurality of networks based on an initial cost value of each grid, each network including a pair of electronic devices to be connected by a wiring path, the optimal wiring path representing a wiring path that passes through a grid with a minimum sum of cost values; b) determining an occupancy state value of each grid, the occupancy state value representing an occupancy state of a grid occupied by the determined optimal wiring path; c) determining a crowded grid, the occupancy state value of the crowded grid exceeding a state threshold; d) increasing the cost value of the crowded grid for at least some of the plurality of networks; and e) redetermining the optimal wiring path for at least some of the plurality of networks based on the increased cost value.

[0008] According to yet another aspect of the present invention, a computer-readable medium storing instructions is provided. When the instructions are executed by a processor, the processor is caused to perform a method according to any embodiment of the present invention.

[0009] When the shortest path of a network passes through a congested grid, since the cost value of the congested grid is increased, the sum of the cost values ​​of the shortest path will also increase. In some cases, the sum of the cost values ​​of the shortest path after the increase may be equal to or even greater than the sum of the cost values ​​of a non-shortest path that does not pass through the congested grid. In this case, because the non-shortest path that does not pass through the congested grid has a smaller or at least the same sum of cost values, the non-shortest path can be re-determined as the best path for the network. Since the new best path no longer passes through the congested grid, the number of wirings passing through the grid is reduced, so that the grid is no longer congested. Therefore, the wiring design for multiple networks can be automatically implemented without causing grid congestion. At the same time, since each best path is still determined based on the minimum sum of cost values, even if it cannot be ensured that the path of each network is the shortest, the overall wiring length for all networks will not increase significantly, and the overall wiring length may even be the shortest under the premise of ensuring that there is no congested grid.

[0010] Still further advantages of the present invention will be appreciated to those of ordinary skill in the art upon reading and understanding the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A block diagram of a grid-based printed circuit board wiring design device 100 according to an embodiment of the present invention is shown;

[0012] Figure 2 Schematic diagrams 200 - 1 and 200 - 2 showing examples of printed circuit board routing designs without considering a crowded grid and considering a crowded grid, respectively; and

[0013] Figure 3 A flowchart of a grid-based printed circuit board wiring design method 300 according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0014] Figure 1 FIG. 1 is a block diagram of a grid-based printed circuit board wiring design apparatus 100 according to an embodiment of the present invention.

[0015] The apparatus 100 includes a path determining unit 110 for determining an optimal routing path for each of a plurality of networks based on an initial cost value of each mesh.

[0016] The length and width of each grid on the printed circuit board can be designed in a certain ratio based on the length and width of the printed circuit board, and usually, each grid can have the same length and width, that is, the grid can be designed as a square. For example, a 40*80 grid array can be designed on a 200mm*400mm printed circuit board, and each grid is a square with a side length of 5mm. However, in some special cases, the grid can also be designed as a rectangle. For example, when the wiring density in the longitudinal and transverse directions is very different, a longer grid side length can be set in the direction with higher density.

[0017] Each grid may be assigned the same initial cost value, for example, each grid is assigned a cost value of 1. Of course, those skilled in the art will appreciate that the initial grid cost value is not limited to 1, and it may be set to any suitable value. In some cases, each grid may also be assigned a different initial value. For example, in the central area of ​​a printed circuit board where congestion is expected to be more likely to occur, the initial cost value of the grid may be set higher, while in the peripheral area of ​​the printed circuit board, the initial cost value of the grid may be set lower. Such a setting may "encourage" wiring to preferentially select grids closer to the periphery, thereby avoiding congestion in the central area where congestion is more likely to occur.

[0018] Each of the multiple networks includes a pair of electronic devices to be connected through a wiring path. For a network, there can be multiple paths of varying lengths to connect its pair of electronic devices. Adding the cost values ​​of all the grids passed by a path can obtain the sum of the cost values ​​of the path. For the case where the initial cost value of each grid is the same, the smaller the sum of the cost values, the smaller the number of grids passed by the path, which means that the path is the shortest. As mentioned above, considering the anti-electromagnetic interference performance and manufacturing cost, the wiring length between a pair of electronic devices should be reduced as much as possible. Therefore, the path with the smallest sum of initial cost values ​​can be determined as the optimal wiring path for the network.

[0019] The optimal routing path can be determined using a variety of cost-based methods, including but not limited to: Dijkstra algorithm, depth-first search algorithm (DFS) algorithm, and A* algorithm.

[0020] In a preferred embodiment, the A* algorithm is used to determine the optimal routing path. For any grid n, the A* algorithm can simultaneously calculate the actual cost value g(n) from the starting point of the path to the grid n and the estimated cost value h(n) from the grid n to the end point of the path, thereby obtaining the total cost value of the path C(n) = g(n) + h(n). Compared with other algorithms that always find the end point from the starting point, the A* algorithm can effectively shorten the time to find the optimal routing path, thereby improving the efficiency of routing design.

[0021] The apparatus 100 further comprises a grid occupation state determination unit 120, which is used to determine an occupation state value of each grid, wherein the occupation state value indicates an occupation state of the grid occupied by the determined optimal routing path.

[0022] After the optimal routing paths are determined for multiple networks, there are now multiple optimal routing paths that have been determined on the printed circuit board, each of which corresponds to a network. It is very likely that there is a grid through which a large number of optimal routing paths pass. For example, the paths from top to bottom and from left to right may all pass through the central area of ​​the printed circuit board. Therefore, in the central area, it is very likely that there is a grid occupied by a large number of paths. For another example, in the corner area of ​​the printed circuit board, there may be a grid that is not occupied by any path or is occupied by only a very small number of paths. In other words, the occupancy state of each grid occupied by the determined optimal routing path is different. The corresponding occupancy state value can be determined for each grid to represent the occupancy state of each grid.

[0023] In one embodiment, the occupancy status value of a grid may simply represent the total number of paths passing through the grid, and both the paths passing through horizontally and vertically are accumulated to generate the occupancy status value of the grid.

[0024] In another embodiment, the occupancy state value of a grid may represent the number of parallel paths passing through the grid. If parallel paths exist only in one direction, only the number of parallel paths in that direction may be determined as an occupancy state value. If parallel paths exist in both directions, the number of parallel paths needs to be calculated for each of the two directions and two occupancy state values ​​are formed.

[0025] In another embodiment, the occupancy value of a grid may represent the number of intersections in the grid. When paths in both directions pass through the grid, paths in different directions will inevitably form intersections in the grid. In this case, the number of intersections may be used as the occupancy value of the grid.

[0026] The apparatus 100 further comprises a crowded grid determining unit 130, which is used to determine a crowded grid. If the occupancy state value of a grid exceeds a predefined state threshold, the grid may be determined as a crowded grid.

[0027] In the case where the occupancy state value of a grid represents the total number of paths passing through the grid, the state threshold is the maximum value of the total number of paths that the grid can accommodate, wherein the maximum value is related to the size of the grid and the line width and line spacing to be routed. In some cases, the user can customize the state threshold based on his experience. In other cases, the state threshold can be automatically determined based on predefined criteria. For example, in the case of a layer of a printed circuit board, the maximum value of the total number of paths that the grid can accommodate is: a / (w+d)+b / (w+d), rounded down, wherein a is the length of the grid, b is the width of the grid (in the case of a square a=b), w is the line width, and d is the line spacing. In the case of a multi-layer printed circuit board, the maximum value of the total number of paths that the grid can accommodate is the calculated value multiplied by the corresponding number of layers.

[0028] The crowded grid determination unit 130 may determine whether the total number of paths passing through a grid is greater than the maximum total number of paths that the grid can accommodate, and if so, determine the grid as a crowded grid.

[0029] In addition, in some cases, it is not enough to simply consider the total number of paths that the grid can accommodate. Crosstalk refers to the mutual interference between different networks on the printed circuit board caused by long parallel wiring, mainly due to the distributed capacitance and distributed inductance between parallel lines. Parallel lines are the main cause of mutual interference between signal lines.

[0030] Therefore, as described above, the occupancy state value of a grid can represent the number of parallel paths passing through the grid. In this case, the state threshold is the maximum number of parallel paths that can be allowed, and similarly, the maximum number of parallel paths is also related to the size of the grid and the line width and line spacing to be routed. In some cases, the user can customize the state threshold based on his experience. In other cases, the state threshold can be automatically determined according to predefined criteria. For example, in the case of a single-layer printed circuit board, the maximum number of parallel paths allowed is: l / (w+d), rounded down, where l is the grid side length in the corresponding direction, w is the line width, and d is the line spacing. In the case of a multi-layer printed circuit board, the maximum number of parallel paths allowed is the calculated value multiplied by the corresponding number of layers. As mentioned above, if there are parallel paths in only one direction, only one occupancy state value needs to be determined for that direction, and therefore, only one state threshold exists. If there are parallel paths in both directions, the occupancy state values ​​need to be determined for the two directions respectively, and accordingly, two state thresholds need to be provided. The two thresholds can be the same or different. For example, when the grid is a square, the side length l in each direction is the same, and the same state threshold can be calculated.

[0031] The crowded grid determination unit 130 can determine whether the number of parallel paths passing through a grid is greater than the maximum number of parallel paths allowed by the grid, and if so, the grid is determined to be a crowded grid. In the case where there are parallel paths in both directions, as long as the number of parallel paths in one direction is greater than the corresponding maximum number of parallel paths, the grid is determined to be a crowded grid.

[0032] In addition to the number of parallel paths, the number of path intersections within the grid is also a factor that needs special consideration. For intersections on a printed circuit board, holes are usually punched in the printed circuit board to utilize the front and back sides of the printed circuit board for routing. It is usually desirable to avoid punching holes as much as possible because punching holes increases the manufacturing cost of the printed circuit board and affects the aesthetics of the printed circuit board.

[0033] Therefore, as mentioned above, the occupancy state value of a grid may represent the number of intersections in the grid. The number of intersections may be limited by considering factors such as manufacturing cost, and the maximum number of allowed intersections may be used as the state threshold.

[0034] The crowded grid determining unit 130 may determine whether the number of intersections in a grid is greater than the maximum number of intersections allowed by the grid, and if so, determine the grid as a crowded grid.

[0035] The apparatus 100 further includes a cost value adjustment unit 140 for increasing the cost value of the congested grid.

[0036] After the congested grid determining unit 130 determines the congested grid, the cost value adjusting unit 140 may increase the cost value of the congested grid accordingly. The purpose of increasing the cost value for the congested grid is to increase the "cost" of the path passing through the congested grid.

[0037] The path determination unit 110 can re-determine the optimal routing paths for multiple networks based on the increased cost value. Since the "cost" for the congested grid is increased, in some cases, the newly determined paths of some networks can be "forced" to bypass the congested grid, thereby reducing the number of paths passing through the congested grid, so that the once congested grid is no longer congested.

[0038] Figure 2 Schematic diagrams 200-1 and 200-2 of printed circuit board wiring design examples without considering crowded grids and considering crowded grids are shown, respectively, to illustrate the principle of how to avoid crowded grids by increasing the cost value of crowded grids, wherein the wiring paths designed for three networks a, b and c are shown. Figure 2 ).

[0039] Figure 2Schematic diagram 200-1 in FIG. 2 shows the shortest paths found for each network without considering whether there is a congested grid. Taking the initial cost value of 1 for each grid as an example, for network a, its shortest path passes through 4 grids, so the sum of its cost values ​​is C(a)=1+1+1+1=4; for network b, its shortest path passes through 7 grids, so the sum of its cost values ​​is C(b)=1+1+1+1+1+1+1=7; and for network c, its shortest path passes through 5 grids, so the sum of its cost values ​​is C(c)=1+1+1+1+1=5. Without considering whether grid congestion occurs and calculating the shortest path of each network separately, a grid A occupied by 3 paths is formed. Assuming that the maximum number of paths that each grid can accommodate is 2, it is obvious that the number of paths 3 accommodated by grid A exceeds the maximum value 2, so grid A is determined to be a congested grid, which causes the wiring design to fail.

[0040] In this case, the cost value of the crowded grid A may be increased, for example, the cost value of the crowded grid A may be increased by 2 to form a new cost value 3.

[0041] At this time, if network b still uses the shortest path determined in schematic diagram 200-1, since the path passes through the congested grid A with a cost value of 3, its total cost value will become C(b)=1+3+1+1+1+1+1=9. Compared with the new path for network b shown in schematic diagram 200-2, although the new path passes through 8 grids and is not the shortest path, these 8 grids do not include grid A, so the sum of the cost values ​​of the path is C'(b)=1+1+1+1+1+1+1+1=8. Since C'(b)<C(b), the new path for network b in schematic diagram 200-2 is determined as the optimal routing path for network b. In this case, as clearly shown in schematic diagram 200-2, there will be no grid occupied by more than 2 paths, thus avoiding the appearance of congested grids and achieving a successful routing design.

[0042] exist Figure 2 In the embodiment, a cost value of 2 is added to the crowded grid. However, those skilled in the art will appreciate that any appropriate increase value may be selected. For example, a cost value of 1 may be added to each crowded grid. Different cost value increments reflect the degree of adding "cost" to the crowded grid, and the user may select an appropriate cost value based on experience according to specific needs.

[0043] return Figure 1 In some cases, increasing the cost of a congested grid only once may not completely resolve the congestion. For example, after increasing the cost of a congested grid, there are still multiple optimal paths of the network passing through the grid, that is, the congestion of the grid has not been alleviated.

[0044] Therefore, in one embodiment, the path determination unit 110, the grid occupancy state determination unit 120, the congested grid determination unit 130, and the cost value adjustment unit 140 may repeatedly perform the process of determining the occupancy state value of each grid, determining the congested grid, increasing the cost value of the congested grid, and re-determining the optimal routing path based on the increased cost value until the congested grid determination unit 130 cannot determine the congested grid, that is, the congestion condition has been completely resolved, so that the routing design can be successfully completed.

[0045] In addition, under some preset conditions, it may be difficult to completely avoid congested grids. For example, when the number of perforations allowed is preset, no matter how much the cost of congested grids is increased, a routing solution without congested grids may not be obtained. Or, even if congested grids can be avoided, the routing of some networks will increase significantly, making the routing cost exceed the cost of increasing the number of perforations.

[0046] Therefore, in one embodiment, after repeatedly performing the process of determining the occupancy state value of each grid, determining the crowded grid, increasing the cost value of the crowded grid, and re-determining the optimal routing path based on the increased cost value for a predetermined number of times, if a routing solution that completely avoids the crowded grid has not been obtained, it is necessary to notify the user of the routing failure so that the user can manually design a detour or choose other solutions to increase the number of punching holes. For example, after repeatedly increasing the cost value for the crowded grid 20 times, there are still crowded grids. At this time, if the cost value continues to be increased, the length of some paths will be greatly increased in order to detour, which means that the routing has failed. The predetermined number of times can be determined according to the size of the printed circuit board and the number of grids. For example, for a larger printed circuit board and more grids, the predetermined number of repetitions is greater.

[0047] In this embodiment, the apparatus 100 further includes a notification unit 150, which is used to notify the user of a wiring failure when the crowded grid determination unit 130 still determines that there is a crowded grid after repeating a predetermined number of times. For example, the notification unit 150 can notify the user of the wiring failure by means of an acoustic or optical alarm, or the notification unit 150 can send an email to the user's computer, mobile phone or other communication device to notify the user of the wiring failure.

[0048] The cost value adjustment unit 140 may increase the cost value of the congested mesh for all networks, or may increase the cost value of the congested mesh for only a part of the networks.

[0049] On a printed circuit board, not every signal line is equally important. For some critical signal lines, interference should be avoided as much as possible. Longer signal lines will inevitably lead to stronger interference. Therefore, for critical signal lines, when determining the best path for them, the shorter path should be ensured as much as possible. In the case of a crowded grid, even if some paths are "forced" to bypass the crowded grid by increasing the cost value of the crowded grid, these critical signal lines should not be "forced" to bypass the crowded grid.

[0050] Therefore, the cost value adjustment unit 140 may not increase the cost value of the congested mesh for the net providing the critical signal line, but may increase the cost value of the congested mesh only for the net providing the non-critical signal line.

[0051] In an optional embodiment, the device 100 may further include a classification unit 160. The classification unit 160 may classify each network as a first non-priority network or a second priority network based on priority. A higher priority indicates that the signal line in the network is more critical, and therefore, a network with a lower priority may be classified as a first non-priority network, while a network with a higher priority may be classified as a second priority network. The classification unit 160 may receive input from a user to classify each network as a first non-priority network or a second priority network based on the user input. Alternatively, the classification unit 160 may automatically classify each network based on some predetermined rules. For example, the classification unit 160 may classify a network including a signal bus for communication (such as a CAN bus and a Flex-Ray bus) and a clock signal line as a second priority network, because these signal lines are the critical signal lines with the highest priority in the design of a printed circuit board. A network including other lines may be classified as a first non-priority network.

[0052] Afterwards, the classification unit 160 may input the classification result into the cost value adjustment unit 140, so that the cost value adjustment unit 140 may increase the cost value of the congested grid only for the first non-priority network and not for the second priority network. As a result, when a congested grid accommodates both the path of the non-priority network and the path of the priority network, it may retain the shortest path of the priority network while "forcing" the path of the non-priority network to take a detour.

[0053] In some cases, if the cost value of the congested grid is not increased at all for the critical network, it may result in a situation where the congested grid cannot be avoided in the end. Therefore, in one embodiment, the cost value of the congested grid can be increased for both the priority network and the non-priority network, but the increment for each network is different.

[0054] Therefore, in an optional embodiment, the device 100 may further include a sorting unit 170. The sorting unit 170 sorts different networks based on priority. The sorting unit 170 may receive input from a user to sort the networks based on the user input. Alternatively, the sorting unit 160 may automatically sort the networks based on some predetermined rules. For example, the signal bus for communication (such as the CAN bus and the Flex-Ray bus), the serial peripheral interface (SPI) signal line and other miscellaneous lines may be sorted, wherein the signal bus for communication has the highest priority, the SPI signal line has the second highest priority, and the other miscellaneous lines have the lowest priority.

[0055] The sorting unit 170 may input the sorting result into the cost value adjustment unit 140, so that the cost value adjustment unit 140 may increase the cost value of the congested grid by different amounts for different networks based on the priority, wherein the increment should be smaller for the network with higher priority. For example, for a network including a signal bus for communication, the cost value of the congested grid is increased by 1; for a network including an SPI signal line, the cost value of the congested grid is increased by 2; and for a network including other miscellaneous lines, the cost value of the congested grid is increased by 3. In this way, the less critical networks can be detoured as much as possible, and at the same time, the critical networks can be detoured when necessary, so as to successfully complete the automatic routing design as much as possible.

[0056] In addition to setting different cost value increments for networks with different priorities, the cost values ​​may also be set differently based on the number of repetitions.

[0057] In a preferred embodiment, at the t+1th repetition, for a certain network, the cost value of a congested grid can be incremented by ΔC t+1 Set to:

[0058] ΔC t+1 =(b+h t )×p t ,

[0059] Where b is the priority of the network, and the higher the priority, the smaller b is; h t is the sum of the occupancy state values ​​of the crowded grid in the previous t repetitions; and p t is the occupancy state value of the crowded grid at the tth repetition. The more repetitions there are, the higher the cost value increment is set, which can reach a feasible solution more quickly, thereby achieving a successful routing design.

[0060] Those skilled in the art can understand that the path determination unit 110, the grid occupancy state determination unit 120, the crowded grid determination unit 130, the cost value adjustment unit 140, the classification unit 160 and the sorting unit 170 can represent hardware processors in a device such as a computer. One or more of these units can be implemented by the same processor or by different processors. Alternatively, the path determination unit 110, the grid occupancy state determination unit 120, the crowded grid determination unit 130, the cost value adjustment unit 140, the classification unit 160 and the sorting unit 170 can also represent software modules in an application for determining an optimal routing path.

[0061] Figure 3 A flowchart of a grid-based printed circuit board wiring design method 300 according to an embodiment of the present invention is shown.

[0062] In 310 , an optimal routing path is determined for each of a plurality of networks based on the initial cost value of each mesh, each network including a pair of electronic devices to be connected by the routing path, wherein the optimal routing path represents a routing path that minimizes the sum of cost values ​​of meshes passed through.

[0063] In 320 , an occupancy state value of each mesh is determined, the occupancy state value representing an occupancy state of the mesh by the determined optimal routing path.

[0064] At 330 , a crowded grid is determined, the occupancy state value of the crowded grid exceeding a state threshold.

[0065] At 340, a cost value of the congested mesh is increased for at least some of the plurality of networks.

[0066] In 350 , the optimal routing path is re-determined for the at least some of the plurality of nets based on the increased cost value.

[0067] It can be understood that the method according to the present invention has the same or similar embodiments as the apparatus according to the present invention. For example, operations 310-350 can be repeatedly performed until it is determined in operation 330 that there is no crowded grid, or it is determined in operation 330 that there is still a crowded grid after a predetermined number of repetitions, at which time method 300 ends in 360.

[0068] Figure 3 The method shown in can be implemented by a processor executing corresponding instructions. The instructions can be stored on any appropriate computer readable medium.

[0069] The above is only for reference Figure 3The illustrated embodiments describe the method of the present invention. It is understood that the operations included in the above embodiments are not restrictive and can be deleted, combined, changed, split and / or recombined as needed to add / modify / delete corresponding functions.

[0070] The apparatus and method of the present invention are described above with reference to various embodiments, wherein the embodiments mentioned may include specific features, structures or characteristics, but not every embodiment necessarily includes the specific features, structures or characteristics. In addition, some embodiments may have some or all of the features described for other embodiments or have no features described for other embodiments.

[0071] The various features of different embodiments or examples can be combined in a variety of ways with some of the features included and other features excluded to adapt to a variety of different applications. The accompanying drawings and the foregoing description give examples of embodiments. It will be appreciated by those skilled in the art that one or more of the elements described can be combined into a single functional element. Alternatively, certain elements can be divided into multiple functional elements. Elements from one embodiment can be added to another embodiment. For example, the order of the processes described herein can be changed and are not limited to the methods described herein. In addition, the operations of any flow chart do not need to be implemented in the order shown; nor do they necessarily need to perform all operations. In addition, those operations that are not dependent on other operations can be performed in parallel with other operations. The scope of the embodiments is by no means limited by these specific examples. Whether or not it is clearly given in the specification, many changes such as differences in the order of operations, product composition, and structure are possible.

Claims

1. A grid-based printed circuit board wiring design device, comprising: a path determination unit, configured to a) determine an optimal routing path for each of a plurality of networks based on an initial cost value of each mesh, each network including a pair of electronic devices to be connected by the routing path, wherein the optimal routing path represents a routing path that minimizes the sum of cost values ​​of meshes passed through; a grid occupancy state determination unit, which is used to b) determine an occupancy state value of each grid, wherein the occupancy state value represents an occupancy state of the grid occupied by the determined optimal routing path; a crowded grid determining unit, configured to c) determine a crowded grid, wherein the occupancy state value of the crowded grid exceeds a state threshold; as well as a cost value adjustment unit, configured to d) increase the cost value of the congested grid for at least some of the plurality of networks, The path determination unit is further configured to e) redetermine the optimal routing path for at least some of the multiple networks based on the increased cost value.

2. The device according to claim 1, wherein: The path determination unit, the grid occupancy state determination unit, the crowded grid determination unit, and the cost value adjustment unit are used to repeatedly perform operations b)-e) until the crowded grid determination unit determines in operation c) that the crowded grid does not exist.

3. The apparatus according to claim 1, further comprising a notification unit, wherein the notification unit is configured to: After the path determination unit, the grid occupancy state determination unit, the congested grid determination unit, and the cost value adjustment unit repeatedly perform operations b)-e) a predetermined number of times, if the congested grid determination unit still determines in operation c) that the congested grid exists, a notification is issued.

4. The apparatus according to any one of claims 1 to 3, further comprising a classification unit, wherein the classification unit is configured to classify each of the plurality of networks into a first non-priority network or a second priority network based on a priority, in, The cost value adjustment unit increases the cost value of the congested mesh only for the first non-priority network, and does not increase the cost value of the congested mesh for the second priority network.

5. The apparatus according to claim 2 or 3, further comprising a sorting unit, wherein the sorting unit is configured to sort the at least some of the plurality of networks based on priority, in, The cost value adjustment unit increases the cost values ​​of the congested grid differently for the at least some networks based on the ranking, wherein a larger cost value is added for a lower priority network and a smaller cost value is added for a higher priority network.

6. The device according to any one of claims 5, wherein: The cost value adjustment unit increases the cost value of each congested grid by ΔC for each network in the at least some networks at the t+1th repetition. t+1 Set to: ΔC t+1 =(b+h t )×p t , Where b is the priority of the corresponding network, where the higher the priority, the smaller b is; h t is the sum of the occupancy state values ​​of the corresponding crowded grid in the previous t repetitions; and p t is the occupancy state value of the corresponding crowded grid at the tth repetition.

7. The device according to any one of claims 1 to 3, wherein: The occupancy state value of each grid is one of the following: the total number of routing paths passing through the corresponding grid, the number of parallel routing paths passing through the corresponding grid, and the number of routing path intersections in the corresponding grid.

8. A grid-based printed circuit board wiring design method comprising: a) determining an optimal routing path for each of a plurality of networks based on an initial cost value of each grid, each network including a pair of electronic devices to be connected by a routing path, wherein the optimal routing path represents a routing path that minimizes the sum of cost values ​​of the grids passed through; b) determining an occupancy state value of each grid, the occupancy state value representing an occupancy state of the grid occupied by the determined optimal routing path; c) determining a crowded grid, wherein the occupancy state value of the crowded grid exceeds a state threshold; d) increasing the cost value of the congested grid for at least some of the plurality of networks; as well as e) redetermining the optimal routing path for the at least some of the plurality of networks based on the increased cost value.

9. The method according to claim 8, further comprising: Operations b)-e) are repeatedly performed until it is determined in operation c) that the crowded grid does not exist.

10. The method according to claim 8, further comprising: After repeatedly performing operations b)-e) a predetermined number of times, if it is still determined in operation c) that the congested grid exists, a notification is issued.

11. The method according to any one of claims 8 to 10, further comprising: classifying each of the plurality of networks as a first non-priority network or a second priority network based on priority, Increasing the cost value of the congested grid for at least some of the multiple networks includes: increasing the cost value of the congested grid only for the first non-priority network, and not increasing the cost value of the congested grid for the second priority network.

12. The method according to claim 9 or 10, further comprising: sorting the at least some of the plurality of networks based on priority, Wherein, increasing the cost value of the congested grid for at least some of the multiple networks includes: increasing the cost value of the congested grid differently for the at least some of the networks based on the ranking, wherein a larger cost value is added for a lower priority network, and a smaller cost value is added for a higher priority network.

13. The method according to any one of claims 12, wherein: Increasing the cost value of the congested mesh for at least some of the plurality of networks includes: At the t+1th repetition, for each of the at least some networks, the cost value of each congested grid is incremented by ΔC t+1 Set to: ΔC t+1 =(b+h t )×p t , Where b is the priority of the corresponding network, where the higher the priority, the smaller b is; h t is the sum of the occupancy state values ​​of the corresponding crowded grid in the previous t repetitions; and p t is the occupancy state value of the corresponding crowded grid at the tth repetition.

14. The device according to any one of claims 8 to 10, wherein: The occupancy state value of each grid represents one of: the total number of routing paths passing through the corresponding grid, the number of parallel routing paths passing through the corresponding grid, and the number of routing path intersections in the corresponding grid.

15. A computer readable medium storing instructions which, when executed by a processor, cause the processor to perform the method according to any one of claims 8 to 14.