Multi-currency call billing system
Patent Information
- Application Number
- CN202610750660.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-05-28
- Publication Date
- 2026-09-22
AI Technical Summary
其中，端点取值法忽略了通话期间汇率的真实波动细节，若通话结束时刻恰好处于两个采样点之间，使用最近采样点的值可能与通话期间的实际平均汇率存在较大偏差；而简单平均法虽然考虑了多个采样点，但无法反映采样点之间的瞬时变化趋势，且计算逻辑较为粗糙
[0014]本公开通过采用黎曼曲面处理模块将离散汇率数据拟合为全局连续光滑的曲线并进行高分辨率预计算存储，能够消除传统离散计费因汇率阶梯跳变导致的相邻通话费用“伪波动”，显著提升计费的公平性与一致性。同时，结合平滑汇率存储模块的查询表机制与计算执行模块的查表累加策略，能够在避免硬件执行复杂实时积分运算的前提下，实现高精度即时汇率获取，可以有效平衡计费精度与系统实时性，降低PBX计费单元的计算负载。
Smart Images
Figure CN122802626A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of communication billing technology, specifically to a multi-currency call billing system. Background Technology
[0002] In multinational corporate communication systems, PBX (Private Branch Exchange) equipment not only handles voice call routing and switching but also manages complex multi-currency real-time billing functions. Typical application scenarios include: Multinational hotel groups: Headquartered in the United States, with hotels worldwide. Each hotel's PBX needs to convert call charges in local currency to US dollars according to the real-time exchange rate for centralized financial accounting at headquarters. Outsourced call centers: Providing services to customers in multiple countries, they need to issue bills based on the customer's country's currency (such as Euro, British Pound, etc.), while the underlying billing system may operate in a single base currency. In these scenarios, the PBX's internal hardware billing unit needs to obtain real-time exchange rates to convert the base call charges from the billing currency to the customer's settlement currency. Currently, there are two main ways for PBX billing units to obtain exchange rate data: one is by the system administrator manually configuring a fixed exchange rate; the other is by using NTP synchronization or scheduled tasks to pull the latest exchange rate data from the cloud's exchange rate application programming interface (API) at fixed intervals. Regardless of the method used, the exchange rate data that ultimately enters the billing logic is essentially a sequence of sampled values at discrete points in time. While existing solutions can achieve basic multi-currency conversion, they present the following significant technical challenges when handling continuous calls across time points and high-concurrency billing: (1) Billing “ladder effect” and lack of fairness caused by discrete data. Since the exchange rate data obtained by the PBX billing unit is usually a discrete exchange rate value pulled from the cloud at fixed periods, the change of exchange rate over time is presented as a “ladder function” rather than a continuous curve in the system. When a long call spans multiple exchange rate update points, the existing technology usually adopts the “endpoint value method” (using the most recent sampled value at the start or end time) or the “simple average method” (calculating the arithmetic mean of discrete values within the interval). Among them, the endpoint value method ignores the details of the actual fluctuation of the exchange rate during the call. If the end time of the call happens to be between two sampling points, the value of the most recent sampling point may deviate significantly from the actual average exchange rate during the call. Although the simple average method considers multiple sampling points, it cannot reflect the instantaneous change trend between sampling points, and the calculation logic is relatively crude. This processing method makes the billing results unable to truly reflect market fluctuations, which can easily lead to customers questioning the fairness of the billing.
[0003] (2) The phenomenon of "pseudo-fluctuations" between adjacent calls seriously undermines billing consistency. When two calls of the same duration and destination occur at times that fall on opposite sides of an exchange rate jump, significant cost differences arise due to the application of different tiered exchange rate values. For example, suppose the exchange rate jumps from 0.92 to 0.93 at 14:31:00. Call A (14:30:20-14:35:20) is charged at 0.92 throughout, while call B (14:31:10-14:36:10) is charged at 0.93 throughout. Although the service costs of the two calls are identical and the start times differ by only 50 seconds, the cost difference is approximately 5%. This difference may be negligible for a single call, but accumulated over hundreds of calls per month, it creates abnormal cost fluctuations that can be perceived by customers. This "pseudo-fluctuation" caused by data dispersion seriously affects the smoothness of the billing system and user trust.
[0004] (3) There is an irreconcilable contradiction between the limited embedded hardware resources and the demand for high-precision computing. The billing unit of the PBX device is a typical embedded system with very limited storage and computing resources. On the one hand, if the step effect is to be reduced by increasing the resolution of exchange rate data, a massive amount of historical exchange rate data points need to be stored, which poses a huge challenge to the storage space of the embedded device. Summary of the Invention
[0005] This disclosure provides a multi-currency call billing system.
[0006] According to one aspect of this disclosure, a multi-currency call billing system is provided, comprising: an exchange rate data acquisition module, used to acquire discrete exchange rate value data and form a discrete exchange rate value sequence containing exchange rate values corresponding to different times; a Riemann surface processing module, used to receive the discrete exchange rate value sequence, and construct a smooth exchange rate curve function based on the exchange rate values corresponding to different times contained in the discrete exchange rate value sequence through a numerical fitting algorithm, and discretize the smooth exchange rate curve function on a time axis to obtain smooth exchange rate values at different times, forming a smooth exchange rate value sequence; a smooth exchange rate storage module, used to construct and store a smooth exchange rate value lookup table based on the smooth exchange rate value sequence, the smooth exchange rate value lookup table recording smooth exchange rate values corresponding to different times; and a calculation execution module, used to record the call start time and call end time when the call is established, and after the call ends, determine the call time period based on the call start time and call end time, and based on the call time period, call the smooth exchange rate value lookup table stored in the smooth exchange rate storage module to query the smooth exchange rate values corresponding to each call time contained in the call time period, and calculate the call fee based on the smooth exchange rate values.
[0007] According to at least one embodiment of the multi-currency call billing system of this disclosure, a smooth exchange rate curve function is constructed using a numerical fitting algorithm based on the exchange rate values corresponding to different times in the discrete exchange rate value sequence. This includes: mapping the exchange rate values corresponding to each time in the discrete exchange rate value sequence to complex plane coordinate points; performing a logarithmic transformation on the imaginary components of the complex plane coordinate points to obtain transformed coordinate points; forming a discrete sampling point sequence based on these transformed coordinate points; constructing a cubic spline interpolation function for each time interval between two adjacent discrete sampling points in the discrete sampling point sequence; and smoothly splicing the cubic spline interpolation functions corresponding to adjacent time intervals to obtain a logarithmic exchange rate curve, such that the logarithmic exchange rate curve satisfies the conditions of continuity of the first and second derivatives at each discrete sampling point; and performing an inverse exponential transformation on the imaginary components of the logarithmic exchange rate curve to obtain a smooth exchange rate curve function in the real time domain.
[0008] According to at least one embodiment of the multi-currency call billing system of this disclosure, the cubic spline interpolation function is expressed as: e(t) = a·t³ + b·t² + c·t + d Where a, b, c, and d represent fitting coefficients, t represents time, and e(t) represents the smoothed exchange rate value at time t.
[0009] According to at least one embodiment of the multi-currency call billing system of this disclosure, the smooth exchange rate storage module uses a circular buffer to store the smooth exchange rate value sequence.
[0010] According to at least one embodiment of the multi-currency call billing system of this disclosure, the circular buffer is configured to retain a smoothed exchange rate value sequence over a preset duration of 24 hours.
[0011] According to at least one embodiment of the multi-currency call billing system of this disclosure, the call charge is calculated based on the smoothed exchange rate value, including: determining the billing enable flag at each moment within the call time period, wherein the billing enable flag is 1 when the call is connected and 0 when the call is muted or waiting; dividing the call time period into multiple time segments, such that each time segment corresponds to only one smoothed exchange rate value; calculating the charge increment within each time segment based on the smoothed exchange rate value corresponding to each time segment, wherein the charge increment is equal to the product of the duration of the time segment, the billing enable flag, and the corresponding smoothed exchange rate value; and summing the charge increments of all time segments within the call time period to obtain the call charge.
[0012] According to at least one embodiment of the multi-currency call billing system of this disclosure, the exchange rate data acquisition module, the Riemann surface processing module, and the smoothed exchange rate storage module are configured to execute during off-peak call periods.
[0013] According to at least one embodiment of the multi-currency call billing system disclosed herein, the system is integrated into the firmware of the digital signal processor or microcontroller of the PBX device and operates as an independent front-end data processing module without changing the original calculation logic and hardware interface of the original billing unit.
[0014] This disclosure employs a Riemann surface processing module to fit discrete exchange rate data into a globally continuous and smooth curve and performs high-resolution pre-computation storage. This eliminates the "pseudo-fluctuations" in adjacent call charges caused by stepwise exchange rate changes in traditional discrete billing, significantly improving the fairness and consistency of billing. Simultaneously, by combining the lookup table mechanism of the smooth exchange rate storage module with the lookup-and-accumulation strategy of the calculation execution module, high-precision real-time exchange rate acquisition can be achieved without requiring complex real-time integration calculations in hardware. This effectively balances billing accuracy and system real-time performance, reducing the computational load on the PBX billing unit. Attached Figure Description
[0015] The accompanying drawings illustrate exemplary embodiments of the present disclosure and, together with the description thereof, serve to explain the principles of the present disclosure. These drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this specification.
[0016] Figure 1 This is a schematic diagram of the structure of a multi-currency call billing system according to one embodiment of the present disclosure.
[0017] Figure 2 This is a flowchart illustrating a method for constructing a smooth exchange rate curve function according to one embodiment of the present disclosure.
[0018] Figure 3 This is a flowchart illustrating a call charge calculation method according to one embodiment of the present disclosure. Detailed Implementation
[0019] The present disclosure will now be described in further detail with reference to the accompanying drawings and examples. It should be understood that the specific examples described herein are for illustrative purposes only and are not intended to limit the scope of the disclosure. Furthermore, it should be noted that, for ease of description, only the parts relevant to the present disclosure are shown in the accompanying drawings.
[0020] It should be noted that, where there is no conflict, the embodiments and features described in this disclosure can be combined with each other. The technical solutions of this disclosure will now be described in detail with reference to the accompanying drawings and embodiments.
[0021] Figure 1 A schematic diagram of a multi-currency call billing system according to one embodiment of this disclosure is shown, which can be integrated into the hardware billing unit of a PBX device. The system mainly includes four functional modules: an exchange rate data acquisition module, a Riemann surface processing module, a smoothed exchange rate storage module, and a calculation execution module.
[0022] (1) Exchange rate data acquisition module, used to acquire discrete exchange rate value data and form a discrete exchange rate value sequence containing the corresponding exchange rate values at different times.
[0023] As one possible implementation, during the process of collecting discrete exchange rate value data, the exchange rate data acquisition module can retrieve the latest exchange rate data from the cloud exchange rate application programming interface through the network communication interface of the PBX device according to a preset time period (e.g., every 60 seconds).
[0024] The collected discrete exchange rate data includes exchange rate values at different times, with each time point corresponding to a unique exchange rate value. Arranging the discrete exchange rate data in chronological order creates a discrete exchange rate value sequence.
[0025] (2) Riemann surface processing module, used to receive discrete exchange rate value sequence, and construct smooth exchange rate curve function by numerical fitting algorithm according to the exchange rate value corresponding to different times contained in the discrete exchange rate value sequence, and discretize the smooth exchange rate curve function on the time axis to obtain smooth exchange rate value at different times, forming a smooth exchange rate value sequence.
[0026] Figure 2 The diagram illustrates a flowchart of a method for constructing a smooth exchange rate curve function according to one embodiment of this disclosure. This method is primarily executed by a Riemann surface processing module, such as... Figure 2 The method shown includes steps S210 to S240. The core objective of this method is to transform discrete, stepped exchange rate sampling points into a globally smooth, continuous exchange rate curve function that conforms to the fluctuation characteristics of exchange rate values.
[0027] In step S210, the exchange rate value corresponding to each time point in the discrete exchange rate value sequence is mapped to a complex plane coordinate point.
[0028] As one possible implementation, the discrete exchange rate value sequence can be represented as (t0, e0), (t1, e1), (t2, e2), ..., (t n , e n ), where t i Representing time i, e i Let i represent the exchange rate value at time i, where i ranges from [0, n] and n represents the maximum time.
[0029] For example, the complex plane coordinate point obtained by mapping the exchange rate value at time i can be represented as: z i = t i + i·e i , where t i For the real part, e i is the imaginary part coefficient, and i is the imaginary unit.
[0030] In step S220, the imaginary components of the complex plane coordinate points are logarithmically transformed to obtain the transformed coordinate points, and a discrete sampling point sequence is formed based on the transformed coordinate points.
[0031] By performing a logarithmic transformation on the imaginary components of the coordinate points in the complex plane, we can obtain the logarithmic exchange rate value y corresponding to each exchange rate value. i =ln(e i This forms a discrete sampling point sequence in logarithmic space: (t0, y0), (t1, y1), (t2, y2), ..., (t n , y n ), y i This represents the logarithmic exchange rate value at time i. Each element (t) in the discrete sampling point sequence i , y i Each of these is a discrete sampling point.
[0032] Because exchange rate fluctuations typically exhibit a multiplicative characteristic (i.e., percentage changes are more stable than absolute value changes), for example, an exchange rate increase from 1.0 to 1.1 (a 10% increase) is economically equivalent to an increase from 10.0 to 11.0 (a 10% increase), but the absolute difference differs across the exchange rate space. A logarithmic transformation converts this multiplicative fluctuation into an additive offset. In logarithmic space, the relative rate of change in the exchange rate becomes linear, making subsequent linear interpolation or spline fitting more stable, better capturing the trend of exchange rate fluctuations, and avoiding fitting biases in low and high exchange rate ranges.
[0033] In step S230, for each time interval between two adjacent discrete sampling points in the discrete sampling point sequence, a cubic spline interpolation function is constructed, and the cubic spline interpolation functions corresponding to adjacent time intervals are smoothly spliced together to obtain the logarithmic exchange rate curve, so that the logarithmic exchange rate curve satisfies the conditions of continuity of the first derivative and continuity of the second derivative at each discrete sampling point.
[0034] Cubic spline interpolation is a piecewise interpolation method that constructs cubic polynomials between adjacent nodes and smoothly splices the cubic spline interpolation functions corresponding to adjacent time intervals. This allows the resulting logarithmic exchange rate curve to remain smooth at the connection points (discrete sampling points), eliminating abrupt changes at inflection points. The final logarithmic exchange rate curve can simultaneously satisfy both first-derivative continuity and second-derivative continuity.
[0035] As one possible implementation, the cubic spline interpolation function can be expressed as: e(t) = a·t³ + b·t² + c·t + d Where a, b, c, and d represent fitting coefficients, t represents time, and e(t) represents the smoothed exchange rate value at time t.
[0036] In step S240, the logarithmic exchange rate curve is transformed from the logarithmic space back to the exchange rate space to obtain a smoothed exchange rate curve function.
[0037] As one possible implementation, during the transformation of the logarithmic exchange rate curve from logarithmic space back to exchange rate space, exponential operations can be performed on the logarithmic exchange rate curve. Since the exponential function is monotonically increasing and infinitely differentiable, and the logarithmic exchange rate curve has a globally continuous second derivative, the transformed smoothed exchange rate curve function also maintains the continuity of its globally continuous second derivative.
[0038] This implementation method can transform discrete, stepped exchange rate data into a continuous, smooth exchange rate curve that conforms to the logic of exchange rate values, thereby eliminating the "step effect" in the original exchange rate data and enabling a reasonable and smoothly transitioning exchange rate value to be obtained at any time.
[0039] As one possible implementation, a high resolution (e.g., one point every 0.1 seconds) can be used when discretizing the smoothed exchange rate curve function on the time axis. This allows the generated smoothed exchange rate value sequence to approximate the continuous smoothed exchange rate curve function with extremely high time granularity. Consequently, during subsequent call billing, the billing execution module does not need to perform complex real-time integration calculations or function evaluations; it only needs to directly look up the instantaneous smoothed exchange rate value at the corresponding moment in memory using the time index. This significantly reduces the computational resource consumption of the PBX hardware billing unit while ensuring billing accuracy. (3) Smooth exchange rate storage module, which is used to construct and store a smooth exchange rate value lookup table based on the smooth exchange rate value sequence. The smooth exchange rate value lookup table records the smooth exchange rate values corresponding to different times.
[0040] (4) Calculation and execution module, which is used to record the start time and end time of the call when the call is established, and after the call ends, determine the call time period based on the start time and end time of the call, and based on the call time period, call the smooth exchange rate value lookup table stored in the smooth exchange rate storage module to query the smooth exchange rate value corresponding to each call time included in the call time period, and calculate the call cost based on the smooth exchange rate value.
[0041] Figure 3 A flowchart illustrating a call charge calculation method according to one embodiment of this disclosure is shown, as follows: Figure 3 The method shown includes steps S310 to S340.
[0042] In step S310, the billing enable flag for each moment within the call period is determined. The billing enable flag is 1 when the call is connected, and 0 when the call is muted or in a waiting state. By using the billing enable flag, the effective billing period and non-billing period (such as muted or waiting states) during the call can be accurately distinguished, ensuring that charges are accumulated only when the user actually occupies communication resources and is in a connected state. This avoids incorrect billing for invalid call durations and improves the accuracy and fairness of billing.
[0043] In step S320, the call time period is divided into multiple time segments, so that each time segment corresponds to only one smoothed exchange rate value.
[0044] In step S330, the cost increment within each time segment is calculated based on the smoothed exchange rate value corresponding to each time segment. The cost increment is equal to the product of the duration of the time segment, the billing enable flag, and the corresponding smoothed exchange rate value.
[0045] In step S340, the cost increments of all time segments within the call period are summed to obtain the call cost.
[0046] This implementation method ensures accurate billing based solely on actual connection duration, while replacing complex real-time integration calculations with a pre-calculated smooth exchange rate sequence. This reduces the computational load on the hardware while eliminating billing "pseudo-fluctuations" caused by discrete exchange rates to improve fairness.
[0047] This disclosure employs a Riemann surface processing module to fit discrete exchange rate data into a globally continuous and smooth curve and performs high-resolution pre-computation storage. This eliminates the "pseudo-fluctuations" in adjacent call charges caused by stepwise exchange rate changes in traditional discrete billing, significantly improving the fairness and consistency of billing. Simultaneously, by combining the lookup table mechanism of the smooth exchange rate storage module with the lookup-and-accumulation strategy of the calculation execution module, high-precision real-time exchange rate acquisition can be achieved without requiring complex real-time integration calculations in hardware. This effectively balances billing accuracy and system real-time performance, reducing the computational load on the PBX billing unit.
[0048] As one possible implementation, the smoothed exchange rate storage module uses a circular buffer to store the smoothed exchange rate value sequence. Utilizing the contiguous memory characteristic of the circular buffer, newly generated smoothed exchange rate values automatically overwrite the oldest historical data in the buffer. This allows for maintaining a window of exchange rate data for the most recent preset duration with a fixed storage space, without requiring dynamic memory allocation or complex data migration.
[0049] In one example, the circular buffer is configured to retain a smoothed sequence of exchange rate values over a preset duration. The preset duration can be set to 24 hours.
[0050] As one possible implementation, the exchange rate data acquisition module, Riemann surface processing module, and smoothed exchange rate storage module can be executed at relatively long intervals (e.g., once a day), such as during off-peak call periods, with each execution taking approximately 1-2 seconds, thus avoiding consuming the computing resources required for real-time call processing. The calculation execution module can be invoked in real-time during each call establishment and duration, performing instant billing by querying the smoothed exchange rate value lookup table, thereby achieving load separation between data processing and real-time billing.
[0051] As one possible implementation, the system is integrated into the firmware of the digital signal processor or microcontroller of the PBX device, and operates as an independent front-end data processing module without changing the original calculation logic and hardware interface of the original billing unit.
[0052] Those skilled in the art should understand that the above embodiments are merely for illustrating the present disclosure and are not intended to limit the scope of the disclosure. Those skilled in the art can make other changes or modifications based on the above disclosure, and these changes or modifications still fall within the scope of the present disclosure.
Claims
1. A multi-currency call billing system, characterized in that, include: The exchange rate data acquisition module is used to collect discrete exchange rate value data and form a discrete exchange rate value sequence containing the corresponding exchange rate values at different times; The Riemann surface processing module is used to receive the discrete exchange rate value sequence, and construct a smooth exchange rate curve function based on the exchange rate values corresponding to different times contained in the discrete exchange rate value sequence through a numerical fitting algorithm. The smooth exchange rate curve function is then discretized on the time axis to obtain the smooth exchange rate values at different times, forming a smooth exchange rate value sequence. The smoothed exchange rate storage module is used to construct and store a smoothed exchange rate value lookup table based on the smoothed exchange rate value sequence. The smoothed exchange rate value lookup table records the smoothed exchange rate values corresponding to different times. The calculation and execution module is used to record the call start time and call end time when the call is established, and after the call ends, determine the call time period based on the call start time and call end time. Based on the call time period, it calls the smoothed exchange rate value lookup table stored in the smoothed exchange rate storage module to query the smoothed exchange rate value corresponding to each call time included in the call time period, and calculates the call cost based on the smoothed exchange rate value.
2. The multi-currency call billing system as described in claim 1, characterized in that, Based on the exchange rate values at different times contained in the discrete exchange rate value sequence, a smooth exchange rate curve function is constructed using a numerical fitting algorithm, including: Map the exchange rate value at each time point in the discrete exchange rate value sequence to a complex plane coordinate point; Logarithmic transformation is performed on the imaginary components of the complex plane coordinate points to obtain the transformed coordinate points, and a discrete sampling point sequence is formed based on each transformed coordinate point. For each time interval between two adjacent discrete sampling points in the discrete sampling point sequence, a cubic spline interpolation function is constructed, and the cubic spline interpolation functions corresponding to adjacent time intervals are smoothly concatenated to obtain a logarithmic exchange rate curve, such that the logarithmic exchange rate curve satisfies the conditions of continuity of the first and second derivatives at each discrete sampling point; and By performing an inverse exponential transformation on the imaginary component of the logarithmic exchange rate curve, a smoothed exchange rate curve function in the real time domain is obtained.
3. The multi-currency call billing system as described in claim 2, characterized in that, The cubic spline interpolation function is expressed as: e(t) = a·t³ + b·t² + c·t + d Where a, b, c, and d represent fitting coefficients, t represents time, and e(t) represents the smoothed exchange rate value at time t.
4. The multi-currency call billing system as described in claim 1, characterized in that, The smoothed exchange rate storage module uses a circular buffer to store the smoothed exchange rate value sequence.
5. The multi-currency call billing system as described in claim 4, characterized in that, The circular buffer is configured to retain a smooth exchange rate value sequence within a preset time period, which is 24 hours.
6. The multi-currency call billing system as described in claim 1, characterized in that, Calculating call charges based on the smoothed exchange rate value includes: Determine the billing enable flag at each time point within the call period, wherein the billing enable flag is 1 when the call is connected and 0 when the call is muted or in a waiting state. The call time period is divided into multiple time segments, so that each time segment corresponds to only one smoothed exchange rate value; Based on the smoothed exchange rate value corresponding to each time segment, the cost increment within each time segment is calculated. The cost increment is equal to the product of the time segment duration, the billing enable flag, and the corresponding smoothed exchange rate value. The call cost is obtained by summing the cost increments of all time segments within the call period.
7. The multi-currency call billing system as described in claim 1, characterized in that, The exchange rate data acquisition module, Riemann surface processing module, and smoothed exchange rate storage module are configured to run during off-peak call periods.
8. The multi-currency call billing system as described in claim 1, characterized in that, The system is integrated into the firmware of the digital signal processor or microcontroller of the PBX device, and operates as an independent front-end data processing module without changing the original calculation logic and hardware interface of the original billing unit.