IP Library Granted Patent US 12665851
Granted Patent B2
US 12665851 · App. 18/657,614 · Granted Jun 23, 2026

Load balancing system and method

Inventors: Ching-Yu Chen (Taipei City, TW); Yi-Ching Chen (Taipei City, TW); Ko-Cheng Liu (Taipei City, TW)
Assignee: Compal Electronics, Inc.
H04L47/125H04L12/2898
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12665851
App. No.
18/657,614
Granted
Jun 23, 2026
Kind
B2
Abstract

A load balancing system and method is provided. The load balancing system includes a core network, CPEs (customer premise equipments), user equipments, and a load balancing equipment. The CPEs receive signals from the core network and connect to each other through a network topology. The user equipments are connected to the CPEs. The load balancing equipment is connected to one CPE, and the load balancing equipment confirms a throughput limit of each CPE and includes a data analysis module and a processing unit. The data analysis module receives network parameters between the CPEs and the user equipments, and calculates premise throughputs of the CPEs. The processing unit generates a route control table according to the premise throughputs. According to the route control table, the CPEs adjusts signal transmission between the CPEs and the user equipments to balance the premise throughputs of the CPEs.

Claims (41)

1 . A load balancing system, comprising:

a core network;

a plurality of CPEs (customer premise equipments), receiving signals from the core network, and connecting to each other through a network topology;

a plurality of user equipments, connected to the plurality of CPEs; and

a load balancing equipment, connected to one of the plurality of CPEs, confirming a throughput limit of each of the plurality of CPEs, and comprising:

a data analysis module, receiving network parameters between the plurality of CPEs and

the plurality of user equipments, and calculating premise throughputs of the plurality of CPEs based on the network parameters;

wherein the network parameters comprise reference signal received powers (RSRP) and signal to interference plus noise ratios (SINR); and

a processing unit, generating a route control table according to the premise throughputs of the plurality of CPEs, wherein according to the route control table, the plurality of CPEs adjusts signal transmission between the plurality of CPEs and the plurality of user equipments to balance the premise throughputs of the plurality of CPEs,

wherein the processing unit of the load balancing equipment performs a first load balancing algorithm to generate the route control table, wherein in the first load balancing algorithm, the processing unit regards a CPE with a premise throughput greater than a first critical throughput as a target CPE from the plurality of CPEs;

the processing unit regards a CPE with a premise throughput less than a second critical throughput as a balancing CPE from the plurality of CPEs; and

the processing unit calculates to make the balancing CPE share a sum of the premise throughput of the balancing CPE and a throughput determined by a difference between the premise throughput of the target CPE and the first critical throughput.

2 . The load balancing system according to claim 1 , wherein the plurality of CPEs communicate with each other through wired connection and/or wireless connection.

3 . The load balancing system according to claim 1 , wherein the load balancing equipment is connected to the one of the plurality of CPEs through wired or wireless connection.

4 . The load balancing system according to claim 1 , wherein the data analysis module calculates a number and capacity of resource blocks and a user throughput required for each of the plurality of user equipments to connect to each of the plurality of CPEs, and calculates the premise throughputs of the plurality of CPEs accordingly.

5 . The load balancing system according to claim 1 , wherein according to a calculation result of the first load balancing algorithm, the processing unit determines whether the premise throughputs of the plurality of CPEs are all less than or equal to corresponding first critical throughputs respectively, and if a determination result is positive, the processing unit outputs the route control table generated by performing the first load balancing algorithm.

6 . The load balancing system according to claim 5 , wherein the processing unit performs a second load balancing algorithm to adjust the route control table if the determination result is negative, wherein in the second load balancing algorithm, the processing unit obtains a QUBO matrix according to the network parameters, adjusts the route control table, and computes the route control table and the QUBO matrix to obtain an overall throughput of the plurality of user equipments by using a quantum annealing algorithm, and after multiple times of iterations, the processing unit obtains a maximum value of the overall throughput and outputs the route control table.

7 . The load balancing system according to claim 1 , wherein the first critical throughput is a product of the throughput limit and a first percentage, the second critical throughput is a product of the throughput limit and a second percentage, and the first percentage is greater than the second percentage.

8 . The load balancing system according to claim 7 , wherein the first percentage is 80%, and the second percentage is 50%.

9 . A load balancing method, comprising steps of:

(a) providing a core network, a plurality of CPEs, a plurality of user equipments and a load balancing equipment, wherein the plurality of CPEs receive signals from the core network and connect to each other through a network topology, the plurality of user equipments are connected to the plurality of CPEs, and the load balancing equipment is connected to one of the plurality of CPEs and confirms a throughput limit of each of the plurality of CPEs;

(b) receiving network parameters between the plurality of CPEs and the plurality of user equipments and calculating premise throughputs of the plurality of CPEs by a data analysis module of the load balancing equipment based on the network parameters;

wherein the network parameters comprise reference signal received powers (RSRP) and signalto interference plus noise ratios (SINR);

(c) generating a route control table according to the premise throughputs of the plurality of CPEs by a processing unit of the load balancing equipment; and

(d) according to the route control table, controlling the plurality of CPEs to adjust signal transmission between the plurality of CPEs and the plurality of user equipments to balance the premise throughputs of the plurality of CPEs;

wherein the step (c) comprises a sub-step of:

(c1) controlling the processing unit of the load balancing equipment to perform a first load balancing algorithm to generate the route control table,

wherein in the first load balancing algorithm, the processing unit regards a CPE with a premise throughput greater than a first critical throughput as a target CPE from the plurality of CPEs;

the processing unit regards a CPE with a premise throughput less than a second critical throughput as a balancing CPE from the plurality of CPEs; and

the processing unit calculates to make the balancing CPE share a sum of the premise throughput of the balancing CPE and a throughput determined by a difference between the premise throughput of the target CPE and the first critical throughput.

10 . The load balancing method according to claim 9 , wherein the plurality of CPEs communicate with each other through wired connection and/or wireless connection.

11 . The load balancing method according to claim 9 , wherein the load balancing equipment is connected to the one of the plurality of CPEs through wired or wireless connection.

12 . The load balancing method according to claim 9 , wherein in the step (b), the data analysis module calculates a number and capacity of resource blocks and a user throughput required for each of the plurality of user equipments to connect to each of the plurality of CPEs, and calculates the premise throughputs of the plurality of CPEs accordingly.

13 . The load balancing method according to claim 9 , wherein the step (c) further comprises sub-steps of:

(c2) according to a calculation result of the first load balancing algorithm, determining whether the premise throughputs of the plurality of CPEs are all less than or equal to corresponding first critical throughputs respectively; and

(c3) if a determination result of the sub-step (c2) is positive, controlling the processing unit to output the route control table generated by performing the first load balancing algorithm.

14 . The load balancing method according to claim 13 , wherein the step (c) further comprises a sub-step of:

(c4) if the determination result of the sub-step (c2) is negative, controlling the processing unit to perform a second load balancing algorithm to adjust the route control table,

wherein in the second load balancing algorithm, the processing unit obtains a QUBO matrix according to the network parameters, adjusts the route control table, and computes the route control table and the QUBO matrix to obtain an overall throughput of the plurality of user equipments by using a quantum annealing algorithm, and after multiple times of iterations, the processing unit obtains a maximum value of the overall throughput and outputs the route control table.

15 . The load balancing method according to claim 9 , wherein the first critical throughput is a product of the throughput limit and a first percentage, the second critical throughput is a product of the throughput limit and a second percentage, and the first percentage is greater than the second percentage.

16 . The load balancing method according to claim 15 , wherein the first percentage is 80%, and the second percentage is 50%.