KR101283673B1 - System and method for detecting access point in smartphone - Google Patents

Abstract

In the present invention, by determining the current state of the device of the smartphone to calculate the optimal parameters in consideration of the energy consumption rate of each device to reduce the energy consumption to search the AP, RSSI information received from the Zigbee device through a high-speed CPU It quickly calculates the presence of AP and performs sensing equipment scheduling through its algorithm using channel information obtained from Zigbee and WIFI. That is, the access point detection system of the smart phone according to the present invention is connected to the Zigbee device, the WIFI device, the Zigbee device and the WIFI device, and includes a coordinator for controlling the Zigbee device and the WIFI device, The coordinator may include a device controller for controlling the Zigbee device and the WIFI device, a counter filter for determining whether an AP exists in the channel through RSSI information received from the Zigbee device, and information obtained from the Zigbee device and the WIFI device. It includes a scheduler for determining the scanning device based on the.

Description

System and method for detecting access point in smartphone

The present invention relates to a system and method for detecting an access point of a smartphone, and more particularly, to a system and method for detecting an 802.11 access point through cooperation of Zigbee and WIFI in a smartphone.

With the widespread adoption of 802.11 Wireless LAN (WI-FI), users can use the wireless Internet with a variety of mobile devices, including laptops and smartphones.

The biggest problem of these mobile devices is the limitation of the battery. Among them, smartphones constantly search 802.11 channels to find 802.11 access points (APs) when using WIFI, and energy consumption increases according to the dynamic mobility of smartphones.

Many researches have been conducted on the method of discovering the AP in the 802.11 infrastructure. Wireless Internet basically assumes a mobile situation. Therefore, the discovery of AP as a move is an essential element for wireless communication. These studies can be largely classified into a method of using only 802.11 standard equipment and an additional equipment in addition to the standard equipment.

In the method using only 802.11 standard equipment, the channel search has been developed to efficiently search for a channel in the absence of an AP or to switch between APs in the presence of multiple APs. A representative study of this type is proactive scan. In this study, the station and the AP periodically switch to the power saving mode (PSM), active scan of another channel for a short time, and then return to the existing channel. A temporary packet drop may occur.

The use of additional equipment is being actively researched as smart phones emerge, and can be classified into signal detection, signature and history, movement monitoring, and dual communication. Each method is briefly described below.

A. Signal Detection

The signal detection method is a method of determining the presence of the AP by analyzing the current channel environment. The energy detection method is divided into WIFI NIC and low power separate equipment. Various researches are in progress. Among the representative ZI-FIs, the beacon signal of the AP is received in Zigbee, and it is primarily filtered by comparing with the airtime of the 802.11 beacon. We use this common multiple folding algorithm to determine the periodicity of the signal and compare the beacon interval to determine the presence of AP. When the signal detection method is classified into two fields, feature detection and energy detection, this method is included in the area of feature detection. The advantage of this feature detection method of ZI-FI is that false positives are reduced compared to energy detection, which can reduce energy consumption due to wrong judgment. However, the use of complex folding methods increases computation time and increases the overall energy consumption as more samples need to be collected. In addition, the increase of false negatives due to the implementation of complex algorithms causes users to miss the opportunity to use the AP.

B. Dual Communication

This method informs the existence of AP by exchanging data between AP and station using low power communication device other than WIFI NIC. This approach is useful when a handoff is required while using delay-sensitive applications such as VoIP because the WIFI NIC can discover other APs while communicating. However, in addition to the WIFI equipment, separate equipment must be constructed, and when the low-power equipment has a short communication distance, there is a disadvantage in that the infrastructure construction is complicated.

C. Signature and History

This method collects AP's geographic location information, specific signature information about peripheral devices, or behavior pattern information of smartphone user in advance, and stores it in the form of database on the server so that AP can access and use this database when needed. Is the way to predict. This method is simple to implement and can be easily applied to various fields. However, since this method requires the collection of various information, it takes time until the database is built to some extent, and the reliability is very poor when the database is not updated well.

D. Movement Monitoring

This method detects an AP by analyzing a user's movement pattern. With the emergence of smartphones with various sensors that can analyze user's movement patterns such as acceleration sensors and gyro sensors, related research is being actively conducted. This method monitors the movement of the smartphone to change the scan pattern of the AP. Among these, WIFIsense adjusts the scanning interval of 802.11 NICs based on the user's moving speed and the density of the AP. The acceleration sensor can be used to stop the user's current status on stationary, walk, and run. Based on the number of APs identified in the previous scan, the AP density is predicted at the next sensing time using an exponential weight moving average. The optimal scan interval until the next scan is calculated using the moving speed information and the AP density information. This process is repeated over time, reducing the energy consumption by adjusting the scan interval accordingly.

In the case of such a movement analysis method, the accuracy of detection itself is high because the 802.11 NIC is used directly in the final judgment, but since the error rate of the sensors analyzing the user's movement is high, a countermeasure for reducing such an error is required.

The present invention has been made on the basis of the above technical background, it is an object of the present invention to provide a method and system for detecting an 802.11 access point of a smartphone that can reduce the detection error and energy consumption.

Another object of the present invention is to provide a method and system for detecting an 802.11 access point through cooperation of Zigbee and Wi-Fi in a smartphone.

In order to solve this problem, the present invention calculates an optimal parameter considering the energy consumption rate of each device by grasping the device state of the current smartphone to reduce the energy consumption for discovering the AP, RSSI information received from the Zigbee device Fast operation is performed through a high-speed CPU to determine the presence of an AP, and sensing equipment scheduling through its algorithm using channel information obtained from Zigbee and WIFI.

That is, the access point detection system of the smart phone according to an aspect of the present invention is connected to the Zigbee device, the WIFI device, the Zigbee device and the WIFI device, the coordinator for controlling the Zigbee device and the WIFI device The coordinator includes a device controller for controlling the Zigbee device and the WIFI device, a counter filter for determining whether an AP exists in a channel through RSSI information received from the Zigbee device, and the Zigbee device and the WIFI device. It includes a scheduler that determines the scanning device based on the information obtained from.

Here, the coordinator may further include a control unit for controlling the device control unit, the counter filter, the scheduler, the coordinator preferably resides in the operating system of the smartphone.

The Zigbee device is idle in the first operation, and operates when the coordinator makes a channel search request, collecting RSSI information at a predetermined time interval for the frequency requested by the coordinator, and collecting the collected RSSI information. When the coordinator operates the WIFI device and transmits a channel scan request, the WIFI device waits in an off state and transmits the channel state to the coordinator. .

According to another aspect of the present invention, there is provided a method for detecting an access point of a smart phone, the method including detecting an access point of a smart phone including a WIFI device and a Zigbee device, including: determining an energy consumption rate of the WIFI device and the Zigbee device; Determining the presence of an access point by calculating RSSI information received from a Zigbee device, receiving channel information from the WIFI device, and determining a scanning device based on information obtained from the Zigbee device and the WIFI device. It includes a step.

According to the present invention, by determining the current state of the device of the smartphone to calculate the optimal parameters in consideration of the energy consumption rate of each device to reduce the consumption of energy for discovering the AP, by using the channel information obtained from Zigbee, WIFI Scheduling of sensing devices through algorithms can reduce false alarms.In the worst case with a false positive ratio of 1, APs can be efficiently used while consuming less energy than using a single device using WIFI or feature detection. Can be detected.

1 is a graph comparing energy consumption rates of ZIFI and WIFI.
2 illustrates a basic concept of a frequency filter used in an access point detection system of a smartphone according to an embodiment of the present invention.
3 is a diagram illustrating a scheduling algorithm of an access point detection system of a smartphone according to an embodiment of the present invention.
4 is a block diagram showing the configuration of an access point detection system of a smartphone according to an embodiment of the present invention.
5 is a diagram illustrating a test environment of an access point detection system of a smartphone according to an embodiment of the present invention.
6 to 9 are graphs showing the results of experiments for evaluating the performance of the access point detection system of the smart phone according to the embodiment of the present invention.

Advantages and features of the present invention and methods for achieving them will be apparent with reference to the embodiments described below in detail with the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. In the present specification, the singular form includes plural forms unless otherwise specified in the specification. It is noted that the terms "comprises" and / or "comprising" used in the specification are intended to be inclusive in a manner similar to the components, steps, operations, and / Or additions.

Now, a method and system for detecting an access point of a smartphone according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

The biggest problem for mobile devices in detecting 802.11 access points (APs) is battery limitations. Among them, smartphones constantly search 802.11 channels to find 802.11 access points (APs) when using WIFI, and energy consumption increases according to the dynamic mobility of smartphones. Many studies have been conducted to find APs using devices with relatively low energy consumption, but the AP detection capability of the devices is inferior to that of 802.11 NICs because the devices are not designed to receive 802.11 beacon signals. Therefore, when using such equipment, users must collect a relatively large number of samples to increase accuracy. This in turn increases the operating time of the device, which consumes a lot of energy. As a result, it may consume more energy and produce more inaccurate detection than the active scan of a typical 802.11 NIC. In order to prevent such inefficiency, the present invention first analyzed the hardware characteristics of each equipment to analyze the energy consumption.

The energy consumption rates of 802.11 NICs and ZigBee devices are shown in Table 1 below.

WIFI ZIGBEE Sleep 40 mW MSP430 Active mode 630 μW Listen 800 mW MSP430 Standby mode 2.31 μW Receive 900 mW CC2420 Receive 41.4 mW Transmit 2,000 mW CC2420 Transmit 36.54mW

Table 1 is based on Maxfor TelosB mote and Atheros chipset WIFI NIC, respectively. The energy consumption of each mode was verified by using a multimeter and compared with the data sheet.

Based on these data, we analyzed the energy consumption of the Feature Detection method through sampling among the existing Zigbee methods. ZIFI was selected as the analysis target. ZIFI assumes that the beacon interval is within 80-120ms, and collects up to eight or more samples to determine the presence of AP through common multiple folding. In ZIFI, the energy consumption rate of each device was actually measured. In the case of ZIFI, one sample was collected for 1.22 seconds, and WIFI sensed 5 seconds every 20 seconds. Using these metrics and the instrument's datasheet, the energy consumption of ZIFI was analyzed mathematically. The result is as shown in Figure 1a.

However, if we reanalyze in terms of miss opportunity, we can see that this situation is unfair to WIFI. That is, the scan interval is 1.22 seconds for ZIFI and 20 seconds for WIFI. In addition, in the case of the sensing time (duration), if the ZIFI is a single value, the WIFI must be sensed once, so the sensing time of the WIFI must be 2.12 ms as in the above data.

Therefore, a new energy consumption analysis was conducted to meet these criteria. In this case, the sensing time of ZIFI is set to 800ms, the minimum value proposed. FIG. 1B is a graph illustrating energy consumption when no booting energy and delay of the WIFI NIC are assumed and the WIFI NIC is turned off when no sensing is performed.

1c assumes that the WIFI NIC is waiting in a power saving mode when no sensing is performed.

The analysis shows that ZIFI consumes more energy than WIFI when feature detection is used at the same miss opportunity. The reason for this is that when the feature detection is performed, the accuracy is improved only when the number of samples to be collected is increased. Therefore, the energy consumption is increased by the increase of the operating time.

Next, we analyzed the correlation between the energy efficiency of smartphones and the scan time of ZigBee according to the energy consumption of each device.

The notation used below is shown in Table 2 below.

Energy (W) Time (sec) Ewas WIFI Active Scan (J) Twi WIFI Scan Interval Ewb WIFI Boot-up Twb WIFI Boot-up Delay Ezr ZIGBEE Receive Tzd ZIGBEE Scan Duration Ezs ZIGBEE Sleep Tzi ZIGBEE Scan Interval

Ew and Ez are defined as the energy that the WIFI NIC and Zigbee use for scanning, respectively. At this time, the Zigbee-based method must satisfy Ew> Ez in order to obtain energy gain compared to the case of one-time active scan using the WIFI NIC. When the WIFI does not scan, it is turned off, and if Ewb and Twb = 0, it can be expressed as Equation 1 below.

If Ewas and Ezr are hardware dependent constants, and Tzi and Twi are defined by the user, Tzd can be represented by the following [Equation 2].

The situation in [Equation 1] above is not realistic because it does not take into account the boot-up delay of the WIFI NIC. Therefore, it is assumed that the WIFI NIC normally waits in PSM and performs an active scan when necessary. It can be represented by [Equation 3] and [Equation 4].

[Equation 5] and [Equation 6] represent the case where the WIFI NIC normally waits in the off state and wakes up when necessary. That is, the bootup delay Twb and the bootup energy Ewb are not zero. Here, it is assumed that Ewb follows the average energy consumption rate per second in the PSM of the WIFI NIC, and Twb is set to 255ms, which is the value obtained through experiments on a self-made test bed. The experimental value was used as the average value after 1000 iterations, the test bed used in the experiment will be described in detail later.

Through the above equations, it is possible to predict the upper limit of Tzd and the resulting energy gain under certain conditions. Considering only the energy consumption of Zigbee and WIFI NIC, the upper limit of Tz for each Zigbee manufacturer is shown in Table 3 below. Here WIFI NIC is based on Atheros chipset. We also assume that Zigbee's bootup delay and energy are very small and can be ignored. The Tzd values in Table 3 differ depending on the hardware of the smartphone. According to an embodiment of the present invention, the energy gain of the corresponding device can be calculated using the above-described conditions according to the equipment situation of the individual smartphone, and the device helps to determine whether the device uses Zigbee.

Device Time (sec) Power saving Power off Power off & Boot up Maxfor telosb 1.0118 0.0614 0.3043 Cross bow telosb 0.8159 0.0495 0.2454 Memsic telosb 0.6324 0.0384 0.1902

According to the above analysis, we propose an optimal detection model considering energy consumption. The goal of this detection model is to minimize scan times for false negatives and ZigBee.

This assumes that a false negative occurs due to a scan by ZigBee, and the user misses the opportunity to access the AP, which can be seen as an adverse effect of limiting the user's access to the assistive device. Because.

Also, if you collect a lot of samples to reduce false positives, the scan time will be longer, which in turn increases the energy consumption. However, if a simple algorithm is used to minimize false negatives or if the detection threshold is sensitively set, there is a problem in that false positives increase significantly compared to the conventional methods.

According to an embodiment of the present invention, this problem is solved through scanning equipment scheduling. The ultimate goal of these algorithms is to assume the worst case scenario so that even if the false positive is close to 1, the search interval is the same as the energy consumption rate is lower than that of a single scan by a single device.

A. Combination of Feature Detection and Energy Detection

To this end, in the access point detection system of a smart phone according to an embodiment of the present invention, a detection model using Zigbee is designed. The method of detecting a signal by sampling the state of a channel can be classified into two fields, feature detection and energy detection. Pros and cons of each are listed in Table 4 below.

Feature Detection Energy detection False positive Low High False Negative High Low Sensing Duration Long Short Energy consumption High Low

In the case of feature detection, false positives are relatively small, resulting in low energy consumption due to false judgments. However, by increasing the operating time by generating false negatives and collecting many samples, the overall energy consumption is large.

Energy detection does not require many samples because it only detects the magnitude of the signal, not the noise, on that channel. Therefore, the overall search time is short, which consumes less energy. In addition, there are few false negatives due to the small loss due to filtering, but the false positives are relatively large. In an environment with such a tradeoff, the two approaches are appropriately mixed to meet the above limitations. That is, energy detection is basically performed to minimize time and reduce energy consumption. However, simple feature detection is performed to minimize the reaction to general noise or other equipment.

B. Counter Filter

In the access point detection system of the smart phone according to the embodiment of the present invention, the counter filter is designed to minimize the false negative while satisfying the above limitation of Tzd. The filter aims to find similar artificial signals based on the beacon's airtime. In addition, the time limit is satisfied by minimizing the number of sample collections and the computational complexity.

The basic concept of this frequency filter is shown in FIG.

The filter is basically based on Zigbee's RSSI measurement and AP's beacon signal characteristics. Zigbee used for channel sampling updates the value of the RSSI register every 32us. This value is then the average of the previous 128us. The airtime of the beacon signal of the AP is about 256us-11720us. Based on this property, when analyzing samples obtained at 128us intervals, two or more consecutive identical values are obtained from the register. In this case, false positives may occur due to noise. This is primarily reduced by averaging the sensed value for 128us and storing it in a register. Secondly, when the sample value is uniformly distributed, i.e., when the continuous value occurs more than two times within the threshold Thr _b , the sample is determined as a signal that is not noise. The maximum value of these RSSI values is regarded as the signal of A. In addition, in order to distinguish the data traffic and the beacon signal, it is recognized as a signal only when the number of sensing periods and airtime consideration frequency is within a certain number. For example, in implementation, Thr _b = ± 2dBm. The operation of the counter filter can be represented by the following pseudo code (Algorithm 1), and Table 5 shows the notation used in the counter filter.

================================================== ===========================

Algorithm 1 Counter Filter

================================================== ===========================

Cmax <= Maximal beacon airtime / Ti-1

Cmin <= Minimal beacon airtime / Ti-1

Cbuf <= 0

Xmax <= 0

Receive X from ZIGBEE

for i from 0 to Tw by 1 do

if Xi < thr _a then

Xi = 0

end if

if Xi > xmax or Sbuf > 0 then

if Xbuf == 0 then

Xbuf = Xi

Cbuf = Cbuf + 1

else

if Xi in Thr _b then

Cbuf = Cbuf

else

if C max≥Cbuf≥Cmin then

Xmax = Xbuf

Xbuf, Cbuf = 0

else

Xbuf, Cbuf = 0

end if

end if

end if

end if

end for

================================================== ===========================

Ti Sensing Interval Tw Window size Cmax upper bound of counter Xi sequential value of RSS Cmin lower bound of counter Xmax output from filter Cbuf buffer of counter Xbuf buffer of RSS Thr _a RSS threshold Thr _b RSS relation threshold

C. Scanning Devices Scheduling

The counter filter described above has a high probability of responding to signals of other 802.11 ad hoc devices or Zigbee devices. Considering the battery limitations of smartphones, increasing false positives leads to unnecessary energy consumption, which reduces the overall operational time.

In the access point detection system of a smartphone according to an embodiment of the present invention, such false positives are resolved through cooperation between the WIFI NIC and Zigbee.

By properly adjusting the scanning schedule between the WIFI NIC and ZigBee, it is designed to consume less energy than in any case, even in extreme situations with very high false positives.

3 and Algorithm 2 show a scheduling algorithm of an access point detection system of a smartphone according to an embodiment of the present invention. A scan interval, which is an interval between scanning slots, is defined as Ti. Every slot is scanned using ZigBee or a WIFI NIC. The initial scan is scanned using ZigBee, and the WIFI NIC waits with the power off. If it is determined that the AP exists, the next scan after Ti operates WIFI (on) and scans using it. The energy and time consumed at this time include the energy that the WIFI performs once and active scan at the time and energy at which the WIFI NIC is booted.

Assuming the worst case, the ratio of ZigBee to WIFI is 1: 1 when a false negative occurs in every ZigBee scan. In this situation, set Tzd so that the average energy consumption per second can gain over a single scan of WIFI or ZigBee.

================================================== ===========================

Algorithm 2 Device Scheduling for Channel Scanning

================================================== ===========================

flag <= ZIGBEE

Zigbee_Power <= OFF

WIFI_Power <= OFF

loop

if flog = ZIGBEE then

Zigbee_Power <= ON

Scan the channel and Send Rsz

Zigbee_Power <= OFF

Rfz <= CounterFilter <= Rsz

if Rfz ==Positive then

flag <= WIFI

else

do nothing

end if

else

WIFI_Power <= ON

Scan the channel and Send Rsw

WIFI_Power <= OFF

if Rsw == Positive then

Associate with AP

Breakloop

else

flag <= ZIGBEE

end if

end if

Wait for Scanning interval

end loop

================================================== ===========================

Table 6 below is a notation used in scheduling.

Rfz Filtering result of ZIGBEE Rsz sensing result of ZIGBEE Rsw sensing result of WIFI

D. False Alarm

False alarms in the system can be divided into false positive and false negative, respectively.

1) false negative: In this system, false negative is defined as the case where the system determines that there is no AP in the vicinity when the smartphone is in the communication range of the AP. The cause of this phenomenon is largely a problem with the Zigbee hardware itself. The problem with the ZigBee hardware itself is that the node itself is not designed to sense WIFI, but rather it is built for 802. 15.4. Since the hardware cannot decode the WIFI packet, the sample of the AP should be collected only by determining whether the wireless channel is BUSY using CCA. That is, the only method is to obtain only the existence of energy of the radio channel and its size information and to properly sample it.

In addition, since Zigbee is less sensitive to 5-10dBm information than 802.11 devices in hardware, the area of detection is reduced compared to WIFI devices. False negative due to dual reception sensitivity is a problem of the hardware itself and cannot be overcome.

Therefore, the access point detection system of the smart phone according to the embodiment of the present invention focuses on minimizing false negatives caused by the algorithm. In an access point detection system of a smart phone according to an embodiment of the present invention, since the feature is determined as the continuity of the airtime and the signal strength of the beacon, and the threshold is sensitively set, the error may be minimized in terms of false negatives. have. On the other hand, false negatives should be judged more strictly.

2) false positive: A false positive is a case where the AP is determined in a situation where the smartphone is located in an area where the AP cannot be actually used. The cause of this phenomenon occurs when the actual sampled value is substantially similar to the target feature when using feature detection.

The sample ZigBee collects contains both time and signal strength information. Through this, only values above threshold are extracted first. Next, determine the signal length and compare it to the airtime of the 802.11 beacon. In this case, the determination is made including the similarity of the signal size. In such a situation, false alarms are 1) when there is a data packet with air time similar to beacon, 2) when the radio characteristics of other equipment is similar to the length of the beacon signal, and 3) when continuous noise occurs beyond the threshold. May occur. As a result, the distribution of equipment other than the AP within a certain area (eg, station, etc.) has the greatest influence on the false positive.

An object of the smart phone access point detection system according to an embodiment of the present invention is to obtain the optimum efficiency when providing a WIFI service using a variety of equipment, such as ZigBee, WIFI, CPU. Each of these three devices has different characteristics. ZigBee uses less energy when using ZigBee, but its accuracy is lowered because its CPU speed decreases and it cannot decode the beacon signal of the AP. In the case of WIFI, the AP signal can be detected quickly and accurately, but the standby power is very large. The main processor of the smartphone has a fast operation speed.

The configuration of the access point detection system of the smart phone according to the embodiment of the present invention designed in consideration of the advantages and disadvantages of each of the equipment is shown in FIG.

As shown in FIG. 4, the access point detection system 100 of the smart phone according to the embodiment of the present invention includes a Zigbee device 110 in charge of sampling, a WIFI device 120 confirming the presence of an AP signal, and And a coordinator component 130 that oversees the entire process of filtering the sampled signal and controlling the WIFI device.

The role of each device is as follows.

A. Zigbee device

The role of the Zigbee device 110 is to measure the RSSI strength of the channel to be searched. It waits for the first time and operates only when the coordinator 130 makes a channel search request. The coordinator 130 collects RSSI information stored in a register of the CC2420 chipset at a predetermined time interval for the requested frequency. This information is then quickly delivered to the coordinator 130.

B. WIFI device

In the access point detection system 100 of the smart phone according to the embodiment of the present invention, the on-off right of the WIFI device 120 belongs to the coordinator 130. The WIFI device 120 normally waits in the off state, and when the coordination of the channel is required, the coordinator 130 operates the WIFI device 120 and sends a channel scan request. Upon receiving this, the WIFI device 120 immediately measures the state of the channel according to the requested information, and transmits it to the coordinator 130. The coordinator 130, which receives it, immediately turns off the WIFI device 120.

C. Coordinator

The coordinator 130 is in charge of all operations of the access point detection system 100 of the smart phone according to the embodiment of the present invention. The coordinator 100 resides in the OS of a mobile device such as a smartphone, and 1) calculates an optimal parameter in consideration of the energy consumption rate of each device by grasping the current state of the device of the smartphone to reduce energy consumption for searching for an AP. 2) Quickly calculate the RSSI information received from the ZigBee device through a high-speed CPU to determine the presence of an AP, and 3) Perform scheduling of sensing equipment through its own algorithm using channel information obtained from ZigBee and WIFI. , 4) Reduce the energy consumption by controlling the operation of the device owned by the smartphone.

Looking at the configuration of the coordinator 130 in more detail with reference to Figure 4, the control unit 132 for controlling the overall situation, the device control unit 134, Zigbee device 110 for controlling the Zigbee and WIFI devices 110, 120 It includes a counter filter 136 for determining whether the AP is present in the channel through the RSSI information received from the scheduler, and a scheduler 138 for determining the scanning device based on the information obtained from the Zigbee and the WIFI device (110, 120).

The test bed of the access point detection system of the smart phone according to an embodiment of the present invention was produced and its performance was evaluated.

The fabricated testbed was implemented using telosB mote with CC2440 chip, onboard Intel chipset wireless LAN card, Intel dual core 1.7GHz processor, 2GB RAM, and ubuntu 9.10.

FIG. 5 illustrates a test environment of an access point detection system of a smart phone according to an embodiment of the present invention. The clear environment is an underground parking lot without interference as shown in FIG. As shown in 5 (b), it is a laboratory with a mixture of equipment.

The false alarm of the access point detection system of the smart phone according to the embodiment of the present invention was evaluated through experiments and mathematical analysis. In the case of false negatives, the results were obtained through actual experiments, and in the case of false positives, the queuing model was used for mathematical analysis.

1) false negative: In order to measure the detection accuracy of the access point detection system of the smart phone according to an embodiment of the present invention to measure one channel in a noise-free environment and to measure for several channels in a noisy environment Two kinds of experiments were conducted.

In the first experiment, the RSSI of the beacon signal was measured by using a test bed to confirm the accuracy of the counter filter, and compared with the measured value by the WIFI NIC. The experimental environment is shown in the underground parking lot as shown in FIG. Experiments were conducted on channels without interference. The AP used Anygate AP with adjustable output power, and the output power was divided into three cases of maximum output (10mW), 25% output, 25% output + radio wave shield. Measurements were made at 2.5m intervals from 15m away to 60m. Using the sample sensed for 150ms at each point, the beacon signal value extracted using the counter filter once and the average value after 10 measurements using the WIFI device were measured. At this time, the accuracy of the filter is higher as the WIFI signal changes and the filtering result is matched.

6 (a) and 6 (b) show experimental results when the output is 100% and 25%, respectively. In general, ZigBee receives about 10dBm weaker signal than WIFI and shows good overall performance considering random change of signal.

In the case of FIG. 6C, obstacles are artificially installed to measure false negatives. At point 13, RSSI was received below the threshold of the counter filter. In this case, since WIFI is normally received, this can be regarded as false negative. At the 15th point about 5m away, WIFI did not receive beacons about 30% (3 times out of 10 times), and since it was not continuously received, wireless LAN connection was impossible. Considering this, the actual false negative area is determined to be about 5m, which is not a filter problem but a reception sensitivity of ZIGBEE hardware itself.

The second experiment was to measure several channels in a mixed environment of many and many devices. Experimental environment is the same as in FIG. In this experiment, the first AP is set to channel 2 and the second AP is set to channel 8 according to the signal strength of the target AP, respectively. There were 14 APs in the vicinity. In this environment, channels were sequentially searched to extract data. Based on this data, it is possible to pinpoint the channel where the AP in good signal condition exists in a noisy environment.

7 is a graph showing experimental results. In the second experiment, the AP was found with 100% accuracy in the specified environment.

2) false positives: The false positives were mathematically analyzed using the queuing model. The cause of false positives is believed to be caused by white noise and interference by other equipment. In practice, since the characteristics of the interfering equipment are different from those of the AP, the density of the equipment does not necessarily affect false positives. However, for the sake of simplicity, we used the worst case where equipment appeared unconditionally false positives. Devices that are expected to interfere can be Bluetooth, Zigbee, or WIFI. The most influential one is 802.11, because it has the largest transmit power and continuously sends data of similar length to the beacon signal.

The basic model is as follows.

이다. 사용자의 활동 반경내 단위면적당 802.11 장비의 수는 ρ이다. 사용자가 이동시 802.11 장비를 만나는 간격은 point poisson process이고, 한 장비 내에 있는 시간은

이며, 이는 exponential distribution을 따른다. 이 때

이고, 여기서 R은 802.11의 송신 범위이다. 위의 가정에 따라 장비에 대한 평균 도달율(arrival rate)과 출발률(departure rate)을 각각

와

로 정의할 수 있다. 이에 따라 사용자가 간섭장비의 송신 영역 안에 있을 확률은 λ/μ이다. 결론적으로 false positive는 다음의 [수학식 7]로 나타낼 수 있다.Your average speed is

to be. The number of 802.11 devices per unit area in the user's radius of activity is ρ. The interval at which a user meets an 802.11 device when moving is a point poisson process.

This follows the exponential distribution. At this time

Where R is the transmission range of 802.11. Based on the above assumptions, the average arrival rate and the departure rate for the equipment, respectively,

Wow

. Accordingly, the probability that the user is in the transmission range of the interfering device is λ / μ. In conclusion, the false positive can be expressed by the following Equation 7.

Assuming R = 250m, ρ = 8.0 * 10 ^-6 . This is the same as when the space between devices in the unit disk environment is about 199m. If R = 200m, it is 160m. Based on this, the change of false positives according to the density is shown in FIG. 8, and the false positives increase linearly with increasing density.

As described above, since the access point detection system of the smart phone according to the embodiment of the present invention is to minimize the false negative and energy consumption when the characteristics of the individual equipment is known, the signal processing is briefly configured to minimize the energy consumption. It was. A feature of the access point detection system of a smartphone according to an embodiment of the present invention is to minimize false negatives and to solve the false positives generated by assisting the WIFI NIC. This is because unnecessary energy consumption increases as the false positive increases. Therefore, in the access point detection system of the smart phone according to an embodiment of the present invention, even in the worst case with a false positive ratio of 1, the energy consumption is less than that of using a single device using WIFI or feature detection. The system was designed.

The energy consumption when the access point detection system of the smartphone according to the embodiment of the present invention is in the worst case is compared with the conventional scheme. In this case, the criteria for energy consumption can be seen as background energy consumption of the smartphone itself, and energy consumption of WIFI and ZIGBEE. Here, only energy consumption of WIFI and ZIGBEE equipment is considered, except that energy consumption due to signal processing is very small compared to basic background energy consumption.

The most influential factor in energy consumption is a false alarm. Energy consumption due to false positive is defined by the following [Equation 8]. Assuming the worst case in Equation 8, if the false positive ratio of the ZIGBEE counter filter is 1, the ratio of scan slots in which WIFI and ZIGBEE operate in the entire system is 1: 1.

In order to evaluate the performance of the access point detection system of the smart phone according to the embodiment of the present invention, it was compared with the energy consumption rates of WIFI and ZIFI having the same scan interval. In the case of WIFI, one scan is based on the energy of waking up the device and performing one active scan.In the case of ZIFI, scanning using ZIGBEE for 800ms and waiting 200ms idle and false alarm value is 0. Assumed The beacon interval of the search target AP is assumed to be 200ms.

While the average energy consumption of ZIFI and WIFI measured based on the same scan interval are 33W and 12.8W, respectively, the access point detection system of the smart phone according to the embodiment of the present invention is 9.5W in a case where the false positive rate is 1 With the lowest energy consumption.

In the WIFI and the access point detection system of the smart phone according to the embodiment of the present invention, each energy consumption according to the change of the false alarm is shown in FIG. 9. In FIG. 9, CO-OP represents a case of an access point detection system of a smartphone according to an embodiment of the present invention.

In the case of ZIFI and WIFI, since the false alarm is assumed to be 0, the energy consumption rate has a constant value of 33W and 12.8W, respectively, and the access point detection system of the smart phone according to the embodiment of the present invention increases as the false positive increases. . In the case of ZIFI, since the value is relatively large, it is omitted from the comparison. As shown in FIG. 9, even in the worst case where the false positive ratio is 1, the access point detection system of the smart phone according to the embodiment of the present invention can be seen that the energy consumption rate is lower than other methods.

Although the present invention has been described above with reference to preferred embodiments, the present invention is not necessarily limited to the above-described embodiments, and various modifications or changes can be made without departing from the spirit and scope of the invention. It is intended that the appended claims cover all such modifications and variations as fall within the true spirit of the invention.

100: access point detection system 110: Zigbee device
120: WIFI device 130: coordinator
132: control unit 134: device control unit
136: Counter Filter 138: Scheduler

Claims (6)

In the access point detection system of a smartphone,
With Zigbee device,
With WIFI device,
It is connected to the Zigbee device and the WIFI device, and includes a coordinator for controlling the Zigbee device and the WIFI device,
Wherein the coordinator comprises:
A device controller for controlling the Zigbee device and the WIFI device;
A counter filter for determining whether an AP exists in a channel through the RSSI information received from the Zigbee device;
And a scheduler that determines a scanning device based on the information obtained from the Zigbee device and the WIFI device. The apparatus of claim 1, wherein the coordinator comprises:
And a control unit for controlling the device control unit, the counter filter, and the scheduler. The apparatus of claim 1, wherein the coordinator comprises:
Access point detection system of the smartphone residing in the operating system of the smartphone. The method of claim 1,
The Zigbee device is idle in first operation,
The access point detection system of a smartphone that operates when the coordinator makes a channel search request and collects RSSI information at a predetermined time interval for the frequency requested by the coordinator and delivers the collected RSSI information to the coordinator. The method of claim 1,
The WIFI device is waiting in the off state,
When the coordinator operates the WIFI device and transmits a channel scan request, the access point detection system of the smart phone to measure the channel state according to the channel scan request and to transmit to the coordinator. In the access point detection method of a smartphone including a WIFI device and a Zigbee device,
Determining an energy consumption rate of the WIFI device and the Zigbee device;
Calculating the presence or absence of an access point by calculating RSSI information received from the Zigbee device;
Receiving channel information from the WIFI device;
And determining a scanning device based on information obtained from the Zigbee device and the WIFI device.

Images