WO2019237751A1

WO2019237751A1 - Method and system for detecting power consumption of mobile application

Info

Publication number: WO2019237751A1
Application number: PCT/CN2019/073516
Authority: WO
Inventors: 伍楷舜; 李雪亮
Original assignee: 深圳大学
Priority date: 2018-06-13
Filing date: 2019-01-28
Publication date: 2019-12-19
Also published as: CN108829598B; CN108829598A

Abstract

Provided are a method and system for detecting the power consumption of a mobile application. The method for detecting the power consumption of a mobile application comprises the following steps: step S1, generating test input and running environment parameters; step S2, running a mobile application under the test input and running environment parameters, and simultaneously collecting power consumption information of the mobile application under the test input and running environment parameters; and step S3, analyzing the power consumption information to recognize whether there is a power consumption problem. According to the present invention, by means of generating the test input and running environment parameters, then simulating special running environments, such as an environment where a network component has a failure, and then running and testing a mobile application in the special running environments, more deeply hidden power consumption problems can be found; and through testing, it is found that 84% of the power consumption problems detected in the present invention are not detected in the prior art, and by solving these power consumption problems, the battery life of a mobile terminal can be prolonged to about twice the current battery life.

Description

Method and system for detecting power consumption of mobile applications

[0001] The present invention relates to a power consumption detection method, in particular to a power consumption detection method for a mobile application, and to a power consumption detection system adopting the mobile application power consumption detection method.

Background technique

[0002] In the past ten years, the growth of the number of users of mobile devices has exploded. At the beginning of 2014, the market share of smart mobile devices exceeded desktop and portable computers with 1.7 billion units; by 2017, the market share of smart mobile devices exceeded _1.9 billion units. With the increasing market share of mobile devices, people ’s use habits of electronic devices have also changed. Even though mobile devices have not completely replaced other electronic devices, users spend more and more time on mobile devices. By 2015, the usage time of mobile devices has exceeded the total of other devices.

[0003] However, the battery power of mobile devices is very limited, and low power consumption is already one of the most important design indicators for mobile applications. In most cases, the battery life of a smartphone in normal use is difficult to exceed 24 hours. We randomly selected 89 high-quality mobile applications and found that more than a third of the applications have serious power consumption problems; these power consumption problems can drain the battery in a few hours, and many users have even decided not to use it permanently. These applications are up.

[0004] At present, most of the research on power consumption optimization of mobile devices is focused on a single component. In fact, although each hardware component is also consuming power, mobile applications control hardware through software. Therefore, if software optimization can be achieved, It can save 50% to 80% of the overall power, which is difficult to achieve with a single hardware optimization. Therefore, if the power consumption problem of mobile applications can be found well, it is actually a fundamental prerequisite to repair the power consumption problem in the future.

Summary of invention

technical problem

Problem solution

Technical solutions

[0005] The technical problem to be solved by the present invention is to provide a mobile application that can more effectively detect whether a mobile application is Whether there is a power consumption detection method for the power consumption problem; and further provides a power consumption detection system using the power consumption detection method of the mobile application.

[0006] In this regard, the present invention provides a method for detecting power consumption of a mobile application, including the following steps:

[0007] Step S1, generating test inputs and operating environment parameters;

[0008] Step S2, run a mobile application under the test input and operating environment parameters, and collect power consumption information of the mobile application under the test input and operating environment parameters;

[0009] Step S3, analyzing the power consumption information to identify whether there is a power consumption problem;

[0010] Wherein, in the step S1, the test input and operating environment parameters are parameters in a state where the wireless signal is shielded; the state where the wireless signal is shielded includes a network fault state and a flight mode state.

[0011] A further improvement of the present invention is that, in the step S1, a test input and a running environment parameter are generated in a random or problem trigger mode.

[0012] A further improvement of the present invention is that the test input and the running environment parameters are generated by automatically generating a random user input sequence, and the user input sequence includes an input sequence corresponding to a touch screen and a swipe screen.

[0013] A further improvement of the present invention is that the problem trigger mode includes at least one of a travel positioning background operation trigger mode, a browser background operation trigger mode, and a multimedia playback failure trigger mode.

[0014] A further improvement of the present invention is that the step S3 includes the following sub-steps:

[0015] Step S301, analyze the power consumption information of the mobile application under the test input and operating environment parameters, and determine whether there is a power consumption problem, otherwise return to step S1, and if so, skip to step S302;

[0016] Step S302, recording the power consumption problem, and analyzing and judging whether the power consumption problem belongs to a new problem trigger mode, if otherwise, return to step S1, and if yes, go to step S303;

[0017] Step S303: Record the power consumption problem in the problem trigger mode.

[0018] A further improvement of the present invention is that in the step S301, analyzing the power consumption information of the mobile application under the test input and operating environment parameters includes analyzing a lock screen state phase, a screen opening phase, an application running phase, The power consumption information is placed in the background phase and the screen-standby phase, and then the power consumption information of the five phases satisfies a preset power consumption range, thereby determining whether there is a power consumption problem.

[0019] A further improvement of the present invention is that the preset power consumption range is set to a normal distribution of power consumption of the test case of the mobile application, and by determining whether the power consumption information of the five stages is equal to the power consumption Normal distribution It is consistent, if otherwise, it is determined that there is a power consumption problem in the foreground operation.

[0020] A further improvement of the present invention is to determine whether there is a power consumption problem running in the background by analyzing the relationship between the average power consumption in the screen-on stage and the average power consumption in the background stage. If the average power consumption in the background stage is more than 40% higher than the average power consumption in the screen-on stage, it is determined that there is a power consumption problem running in the background.

[0021] A further improvement of the present invention is to determine whether there is a power consumption problem of running in the background by analyzing the relationship between the average power consumption in the screen-standby state phase and the average power consumption in the lock-screen state phase. If the average power consumption in the standby state of the screen is more than 50% higher than the average power consumption in the state of the lock screen, it is determined that the mobile application has a problem of power consumption without sleep.

[0022] The present invention also provides a power consumption detection system for a mobile application, which adopts the power consumption detection method for a mobile application as described above.

The beneficial effects of the invention

Beneficial effect

[0023] Compared with the prior art, the present invention has the beneficial effects of: simulating a special operating environment such as a network component failure by generating the test input and operating environment parameters, and then operating under this special operating environment And testing mobile applications, we can find more hidden deep power consumption problems; after testing, it is found that 84% of the power consumption problems detected by the present invention are not detected by the prior art, and these power consumption problems are repaired , Can extend the battery life of smart terminals such as mobile phones to about twice the existing technology.

Brief description of the drawings

BRIEF DESCRIPTION OF THE DRAWINGS

[0024] FIG. 1 is a schematic diagram of a work flow according to an embodiment of the present invention;

[0025] FIG. 2 is a schematic diagram of analyzing power consumption information of a mobile application in five stages according to an embodiment of the present invention.

Invention Examples

Embodiments of the invention

[0026] The preferred embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0027] As shown in FIG. 1, this example provides a method for detecting power consumption of a mobile application, including the following steps: [0028] Step S1, generating test inputs and operating environment parameters;

[0029] Step S2, running a mobile application under the test input and operating environment parameters, and collecting power consumption information of the mobile application under the test input and operating environment parameters;

[0030] Step S3, analyzing the power consumption information to identify whether there is a power consumption problem;

[0031] Wherein, in the step S1, the test input and operating environment parameters are parameters in a state where the wireless signal is shielded; the state where the wireless signal is shielded includes a network fault state and a flight mode state.

[0032] The applicant randomly selected 89 mobile applications in the mobile phone application market, and read their development documents, and found that 94.8% of power consumption problems were caused under special operating environments, such as network component failures. Happening. That is, under special circumstances, power consumption problems of mobile applications can be detected more effectively. In step S1 described in this example, test input and operating environment parameters are generated by using a random or problem trigger mode.

[0033] In this example, during the data acquisition phase of steps S1 and S2, a pressure test is applied to the mobile application, that is, a large number of test cases are used to test the target mobile application. For each test case, its test input and operating environment are different. These test inputs and running environments may be randomly generated, or generated according to the problem trigger mode, as described below.

[0034] This example uses Monkey to automatically generate a random user input sequence. The Monkey is an Android application testing tool developed by Google. The user input sequence includes the corresponding sequences for touch screen and swipe screen respectively. At the same time, this example also designed two special operating environments, namely Network Fail and Flight Mode. In the Network Fail operating environment, the wireless network signals are extremely weak, and the transmission delay of the data packet is 451

ms, download speed 21KB / S; and under normal operating environment, the data packet transmission delay is 36 ms, and the download speed is 3.2MB / S. Therefore, in this example, the smart mobile terminal such as a mobile phone is placed in a wireless signal shielding room to implement the test, and then it can be run and tested in the operating environment of Network Fail and Flight Mode. The software layer is closed.

[0035] That is, in this example, a test input and a running environment parameter can be generated by automatically generating a random user input sequence, and the user input sequence includes an input sequence corresponding to a touch screen and a swipe screen.

[0036] The problem trigger mode described in this example includes at least one of a travel positioning background operation trigger mode, a browser background operation trigger mode, and a multimedia playback fault trigger mode. Wherein, the travel positioning runs in the background The implementation of the trigger mode includes the following steps: Turn on the GPS component in the mobile application, place the mobile application in the background, and then make the screen go out automatically. The implementation of the browser background trigger mode includes the following steps:

The plug-in webpage will run the browser in the background; that is, the webpage requires a flash plug-in program. The implementation of the multimedia playback fault trigger mode includes the following conditions: When playing a streaming media, a network fault occurs.

[0037] The problem trigger mode described in this example refers to common test inputs and operating environment parameters that can trigger power consumption problems. For example, for streaming media applications, the network suddenly fails during streaming media playback. This situation (problem trigger mode) can easily trigger power consumption problems.

[0038] It should be particularly pointed out that the operating environment parameters described in this example refer to the special operating environment when the application is running, including situations where the signals of WiFi and cellular networks are extremely weak, situations where GPS signals are extremely weak, and mobile devices. The failure of the WiFi and cellular network components and the failure of the GPS components, etc., are parameters in a state where the wireless signal is shielded.

[0039] As shown in FIG. 1, step S3 in this example includes the following sub-steps:

[0040] Step S301, analyze the power consumption information of the mobile application under the test input and operating environment parameters, and determine whether there is a power consumption problem, if not, return to step S1, and if yes, skip to step S302;

[0041] Step S302, recording the power consumption problem, and analyzing and judging whether the power consumption problem belongs to a new problem trigger mode, if not, return to step S1, and if yes, go to step S303;

[0042] Step S303: Record the power consumption problem in the problem trigger mode.

[0043] The problem trigger mode described in this example refers to a process that may trigger a power consumption problem, and includes at least one of a travel positioning background operation trigger mode, a browser background operation trigger mode, and a multimedia playback fault trigger mode. For specific mobile applications, this example designs a specific input sequence to apply the corresponding operating environment to achieve the corresponding problem trigger mode.

[0044] The criterion for determining the problem trigger mode is: The problem trigger mode has triggered one or more power consumption problems in the test. This guarantees that the problem trigger mode is more likely to trigger power consumption problems than normal, because it has already triggered one or more power consumption problems. And the problem trigger mode can be implemented in the test case, because the test case is based on user input and operating environment. For those power consumption issues that need to be modified by source code, the test case in this example cannot be covered. . [0045] After the test input and the running environment are determined, this example will run the application software with the input sequence under the running environment. During the test, the power consumption information of the device is recorded in real time using the power consumption detector.

[0046] Step S3 in this example is a step in the data analysis phase. At this stage, this example will determine whether the use case has a power consumption problem by analyzing the power consumption information. If there is a power consumption problem, record the power consumption problem and analyze whether the problem is triggered by the new problem trigger mode; if the power consumption problem is triggered by the new problem trigger mode, record the power consumption problem Problem trigger mode. The recorded content includes the user input information and operating environment information required to trigger the power consumption problem. If no power consumption problem is identified or the problem is not triggered by the new "problem trigger mode", the entire test is restarted. Process.

[0047] If no power consumption problem or a problem not triggered by the new problem trigger mode is identified in step S3, it is preferred to automatically restart the entire detection process, that is, return to step S1 to facilitate real-time monitoring of the running status of the mobile application .

[0048] As shown in FIG. 2, in step S301 in the example, analyzing the power consumption information of the mobile application under the test input and operating environment parameters includes analyzing a lock screen state (PRE-OFF) and turning on the screen. Phase (I DLE), application operation phase (EXECUTION), placed in the background phase (BACKGROUND) and the screen standby state phase (SCREEN-OFF) a total of five phases of power consumption information, and then judge the power of the five phases Whether the power consumption information respectively meets a preset power consumption range, and then determine whether there is a power consumption problem.

[0049] The lock screen state phase in this example is also referred to as PRE-OFF, which refers to the stage where the mobile phone is in the lock screen state. When the user turns on the screen, the phone enters the IDLE phase, that is, the screen phase is turned on. When the user opens the target application and runs the application with corresponding user input under the corresponding running environment, the user enters the EXECUTION phase, that is, the application running phase. After the EXECUTION phase, we put the application in the background by entering the "HOME" key. At this time, the test case entered the BACKGROUND phase, which is the background phase. When the screen goes through a standby state, the screen is automatically screened. After the screen is closed, the use case enters the final phase, the SCREEN-OFF phase, which is the screen standby mode.

[0050] This example divides the test cases into five stages to help us identify three types of power consumption problems: foreground operation problems, background operation problems, and no sleep problems.

[0051] In order to identify the foreground operation problem, this example considers the power consumption of all test cases of each application as a normal distribution, and marks test cases that are two standard deviations larger than the average power consumption as possible power consumption problems. Use example. Finally, this example analyzes these test cases to determine if they really have power issues.

[0052] That is, the preset power consumption range is set to a normal distribution of power consumption of the test case of the mobile application, and by determining whether the power consumption information of the five stages is consistent with the normal distribution of power consumption It is consistent, if otherwise, it is determined that there is a power consumption problem in the foreground operation.

[0053] This example determines whether there is a power consumption problem running in the background by analyzing the relationship between the average power consumption in the screen-on stage and the average power consumption in the background stage. If the average power consumption is more than 40% higher than the average power consumption in the screen-on phase, it is determined that there is a power consumption problem running in the background. In practical applications, the value of 40% can be adaptively adjusted according to customer needs, such as the first power consumption threshold set by the user. When the average power consumption in the background stage is greater than the screen opening stage, If the average power consumption is as high as the first power consumption threshold, it is determined that there is a power consumption problem running in the background.

[0054] This example determines whether there is a power consumption problem running in the background by analyzing the relationship between the average power consumption in the screen-standby state phase and the average power consumption in the lock-screen state phase. When the average power consumption in the state phase is more than 50% higher than the average power consumption in the lock screen state phase, it is determined that the mobile application has a power consumption problem without sleep. In practical applications, this value of 50% can be adaptively adjusted according to customer needs, such as a second power consumption threshold set by the user. When the average power consumption during the standby state of the screen is greater than the lock screen, If the average power consumption in the state phase is as high as the second power consumption threshold, it is determined that there is a power consumption problem without sleep.

[0055] This example also provides a power consumption detection system for a mobile application, which uses the power consumption detection method for a mobile application as described above.

[0056] In summary, this example simulates a special operating environment such as a network component failure by generating the test inputs and operating environment parameters, and then running and testing mobile applications in this special operating environment can find more Hidden deeper power consumption problems; after testing, it is found that 84% of the power consumption problems detected by the present invention are not detected by the prior art, and repairing these power consumption problems can replace the battery of smart terminals such as mobile phones The battery life is extended to about twice that of the prior art.

[0057] The above content is a further detailed description of the present invention in combination with specific preferred embodiments, and it cannot be considered that the specific implementation of the present invention is limited to these descriptions. For those of ordinary skill in the technical field to which the present invention pertains, without deviating from the concept of the present invention, several simple deductions or replacements can also be made. All should be regarded as belonging to the protection scope of the present invention.

Claims

Claim

[Claim 1] A method for detecting power consumption of a mobile application, comprising the following steps:

Step S1, generating test inputs and operating environment parameters;

Step S2: Run the mobile application under the test input and operating environment parameters, and collect power consumption information of the mobile application under the test input and operating environment parameters; Step S3, analyze power consumption information to identify whether power exists Power consumption problem; wherein, in the step S1, the test input and operating environment parameters are parameters in a state where the wireless signal is shielded; the state where the wireless signal is shielded includes a network fault state and a flight mode state.

[Claim 2] The method for detecting power consumption of a mobile application according to claim 1, wherein in step S1, a test input and a running environment parameter are generated by a random or a problem trigger mode.

[Claim 3] The method for detecting power consumption of a mobile application according to claim 2, wherein a test input and a running environment parameter are generated by automatically generating a random user input sequence, and the user input sequence includes a touch screen and The input sequence corresponding to the swipe.

[Claim 4] The method for detecting power consumption of a mobile application according to claim 2, wherein the problem trigger mode includes a travel positioning background operation trigger mode, a browser background operation trigger mode, and a multimedia playback fault trigger mode. At least one.

[Claim 5] The power consumption detection method for a mobile application according to any one of claims 1 to 4, wherein the step S3 includes the following sub-steps:

Step S301, analyzing the power consumption information of the mobile application under the test input and operating environment parameters, and determining whether there is a power consumption problem, if not, return to step S1, and if yes, skip to step S302;

Step S302, recording the power consumption problem, and analyzing and judging whether the power consumption problem belongs to a new problem trigger mode, if not, return to step S1, and if yes, go to step S303; step S303, the power consumption problem Record to the problem trigger mode.

[Claim 6] The method for detecting power consumption of a mobile application according to claim 5, characterized in that, in the step S301, analyzing the mobile application under the test input and operating environment parameters The power consumption information includes analyzing power consumption information in five phases including the lock screen state phase, the screen opening phase, the application running phase, the background phase and the screen standby status phase, and then determining whether the power consumption information of the five phases is The preset power consumption ranges are respectively satisfied, and then it is determined whether there is a power consumption problem.

[Claim 7] The method for detecting power consumption of a mobile application according to claim 6, wherein the preset power consumption range is set to a normal distribution of power consumption of a test case of the mobile application, Whether the power consumption information of the five stages described is consistent with the normal distribution of power consumption; if otherwise, it is determined that there is a power consumption problem running in the foreground.

[Claim 8] The power consumption detection method for a mobile application according to claim 6, characterized by analyzing the relationship between the average power consumption in the screen-on stage and the average power consumption in the background stage To determine whether there is a power consumption problem in the background operation, if the average power consumption in the background phase is more than 40% higher than the average power consumption in the screen-on phase, it is determined that there is a power consumption problem in the background operation.

[Claim 9] The method for detecting power consumption of a mobile application according to claim 6, characterized in that, by analyzing the average power consumption in the standby state phase of the screen and the average power consumption in the lock screen phase To determine whether there is a power consumption problem running in the background. If the average power consumption in the standby state phase of the screen is more than 50% higher than the average power consumption in the lock screen phase, it is determined that the mobile application has no power consumption. Sleep power issues.

[Claim 10] A power consumption detection system for a mobile application, characterized by using the power consumption detection method for a mobile application according to any one of claims 1 to 9.