KR100904342B1 - Method for protecting slots and expecting performance measures in unslotted CSMA/CA network - Google Patents

Abstract

The present invention relates to a method for predicting ACK protection and a performance measure of each device in a network configuration using an unslotted CSMA / CA scheme of IEEE 802.15.4. It is designed to protect the ACK by designing longer than the switching time, which is the time until transmission, and predicts the optimal number of devices and battery life in such a network environment. A method for protecting an ACK of each device in a network configuration using an unslotted CSMA / CA scheme of IEEE 802.15.4, wherein the CCA execution time of the device is less than the switching time, which is a time from data reception to ACK transmission. The long design prevents packet collisions between devices. In addition, the performance scale prediction method of the present invention is that the packet used in the structure of the network in which the CCA execution time of the device is longer than the switching time, which is the time from data reception to ACK transmission, the Poisson (λ) rate (rate) a first process generated through a process, and a process of calculating a packet transmission probability α, a backoff time period, an expected delay E [D], and a packet loss probability P _loss after the device performs CCA. And a third step of calculating the number of devices constituting an optimal network in consideration of the packet transmission probability α, the expected delay E [D], and the packet loss probability P _loss .

IEEE 802.15.4, slot, nonslot, beacon, non-beacon, WPAN

Description

Method for protecting slots and expecting performance measures in unslotted CSMA / CA network

1 is a diagram showing a conventional operation timing diagram showing a collision state between devices.

2 is a timing diagram illustrating a design of a CCA for protecting an ACK in an unslotted CSMA / CA scheme in IEEE 802.15.4 according to an embodiment of the present invention.

3 is a diagram illustrating a CCA performance and channel occupancy of each device according to an embodiment of the present invention.

4 is a diagram illustrating an optimal number of devices and a battery life in a network when a CCA execution time is designed to be longer than a switching time from data reception to ACK transmission in an unslotted CSMA / CA scheme according to an embodiment of the present invention. It is a flowchart showing the process of calculating the performance measure such as.

5 is a diagram illustrating a transmission in a cycle period after performing a backoff count according to an embodiment of the present invention.

6 is a graph showing a calculation result and a simulation result according to an embodiment of the present invention.

The present invention relates to a method for predicting ACK protection and a performance measure of each device in a network configuration using an unslotted CSMA / CA scheme of IEEE 802.15.4.

The IEEE 802.15.4 technology standard is showing rapid growth in technology development and market formation compared to other wireless networking standards such as Bluetooth, Wi-Fi, etc., which are struggling in the market. The ISO layer in Low Rate-Wireless Personal Area Networks (LR-WPAN), such as Zigbee, is a physical PHY layer, a medium access control MAC layer, and an upper layer of the network (NWK) layer, applications. It consists of the support (APS) layer, security and application (APL) layer, among which the IEEE 802.15.4 technical standard deals with the PHY layer and the MAC layer.

As a result, the IEEE 802.15.4 technical standard is a low-speed wireless short-range network (LR-WPAN) for low complexity, low power, low speed wireless data connection in processing the physical layer PHY layer and data collision prevention technology MAC layer It becomes standard technology of.

As described above, the basic protocol of LR-WPAN follows the standard defined in IEEE 802.15.4 and uses the CSMA / CA scheme for channel allocation as in IEEE 802.11 and IEEE 802.15.3 based wireless networks. The operation concept of the CSMA / CA algorithm is that IFS (Inter-Frame-Space) that defines the time between frames. In other words, MAC has a period of IFS for processing data received from PHY. This IFS is divided into SIFS (Short IFS) and LIFS (Long IFS) depending on the length of the data.

The CSMA / CA scheme is divided into two types depending on whether slots are used. It is divided into a 'beacon mode' using a slotted CSMA / CA and a non-beacon mode using an unslotted CSMA / CA.

Slotted CSMA / CA is used in a beacon capable network in such a way that a backoff slot is allocated at the same time as the beacon frame transmission. Before sending data, the device waits for a random number based on synchronized backoff slots, and when the channel is busy, the device waits for any number of backoff slots. waiting for a slot). Here, the execution of the Clear Channel Assessment (CCA) is required to protect the ACK packet.

In contrast, unlike the beacon mode of the slot-based CSMA / CA, the unit time of the backoff slot is synchronized in the non-beacon mode of the non-slot-based CSMA / CA. The following problems occur.

That is, as shown in FIG. 1, the Clear Channel Assessment (CCA) implementation time (T _ccA ) is less than the turnaround time from the data reception state to the ACK (acknowledgement) transmission state. short. Therefore, while the first device 11 finishes the data transmission and waits for the ACK from the PAN coordinator, when the other device 12 (the second device) performs the CCA, the first device does not detect that it is in communication. Immediately after the data is transmitted, this leads to a collision with the ACK for the first device 11 and eventually causes both the first device 11 and the second device 12 to fail to transmit data. Occurs.

In order to solve the above problems, the present invention proposes an ACK protection scheme in the non-slot-based (non-beacon mode) CSMA / CA of IEE802.15.4. In addition, a method of calculating the number of devices for configuring an optimal network and a method of predicting battery life of the device are presented.

The ACK protection method of the present invention is a method for estimating the performance measure of each device in a network configuration using an unslotted CSMA / CA scheme of IEEE 802.15.4, wherein the CCA execution time of the device is determined from data reception. It is designed to be longer than the switching time, which is the time until ACK transmission, characterized in that it prevents packet collision between devices. In addition, the device discards the generated packet when the number of CCA executions that fail due to channel busy becomes the number of 'CCA execution maximum value +1'.

In addition, in the method of predicting the performance measure of each device in the network configuration by the unslotted CSMA / CA scheme of IEEE 802.15.4, the performance measure prediction method of the present invention, the CCA execution time of the device In the network designed to be longer than the switching time, which is the time from data reception to ACK transmission, a first process in which a packet used in a structure is generated through a Poisson process having a λ rate, and after the device performs CCA A second step of calculating a packet transmission probability α, an expected delay E [D] having a backoff time period, a packet loss probability P _loss , the packet transmission probability α, an expected delay E [D], And a third process of calculating the number of devices constituting an optimal network in consideration of the packet loss probability P _loss .

에 의해 산출된다.The packet transmission probability α is according to claim 1, wherein T _CCA is CCA execution time, T _TX is packet transmission time, T _ACK is ACK transmission time, P _loss is packet loss probability, and E [D] is backoff time. Given an expected delay, which is a backoff time period,

Calculated by

The expected delay E [D] is M, the maximum value of the backoff performance (4 as a default), T _CCA is the CCA execution time, σ is the length of the backoff slot, and Wj is min {2 ^j W _min , W _max } Where W _min = 2 ^BEmin , W _max = 2 ^BEmax , and BE _min and BE _max represent the minimum backoff index (macMinBE) and the maximum backoff index (macMaxBE), respectively.

Calculated by

에 의해 산출된다.The packet loss probability P _loss When M is the backoff performance maximum value (4 as the default), when a CCA attempt fails at the point M + 1,

Calculated by

Further, in the third process, when generating a packet through a Poisson process having a λ rate, each packet transmission success probability α according to the number of devices for each λ rate, and an expected delay E [D], after calculating the packet loss probability P _loss , the device number corresponding to the network configuration environment is selected.

In addition, the performance measure prediction method of the present invention, the energy consumption per unit time (E _TX ) when transmitting, the energy consumption per unit time (E _RX ) when receiving, the energy consumption per unit time (E _CCA ) when performing _CCA The fourth step of calculating the energy consumption per unit time in one device in consideration of the energy consumption per unit time (E _idle ) and each operation time in the idle state, and the energy amount of the battery of the device The method may further include a fifth process of calculating the battery life time by dividing by the energy consumption per unit time.

The energy consumption per unit time, E _TX , E _RX , E _CCA , E _idle is the energy consumption per unit time (per millisecond) required when transmitting (TX), receiving (RX), CCA, idle, T _CCA is the CCA execution time, T _TX is the packet transmission time, T _ACK is ACK transmission time, E [D] is the estimated delay (delay) back-off time period (backoff time period), N ^CCA is successful transmission or loss When we show the mean number of CCA until there is

Calculated by

에 의해 산출된다.When N ^CCA is a maximum value of M performed backoff (4 as a default), when a CCA attempt fails at M + 1,

Calculated by

Hereinafter, the detailed description of the preferred embodiments of the present invention will be described with reference to the accompanying drawings. In the following description of the reference numerals to the components of the drawings it should be noted that the same reference numerals as possible even if displayed on different drawings.

2 is a timing diagram illustrating a design of a CCA for protecting an ACK in an unslotted CSMA / CA scheme in IEEE 802.15.4 according to an embodiment of the present invention.

In general, a turnaround time, which is an interval from a data reception state to an ACK transmission state, is 220 [usec], and a CCA execution time T _ccA is known as 128 [usec]. As described above, since the CCA execution time T _ccA is shorter than the switching time, data collision may occur between devices, thereby causing data transmission failure. For example, as shown in FIG. 1, if the second device performs CCA during the switching time of the first device, the second device does not detect that the first device is in communication and thus detects the channel idle state. As a result, it sends its own data packet immediately after the CCA. This leads to a collision with the ACK of the first device, and eventually, both the first device and the second device fail to transmit data.

In order to solve this problem, the present invention is designed to make the CCA execution time longer than the switching time which is the time from data reception to ACK transmission. Therefore, other devices performing CCA during the switching time of a specific device can detect the channel busy (busy) to slow down the data transmission of the other device, so that no data collision occurs between the devices. 2, even when the second device 12 performs the CCA during the switching time of the first device 11, the CCA execution time of the second device 12 is the first device 11. Since the second device 12 detects the channel busy due to the ACK of the first device 11, the second device 12 delays the transmission of its data packet so that the second device 12 Data collision between the second devices 12 can be prevented.

In the structure in which the CA execution time is longer than the switching time, which is the time from data reception to ACK transmission, as described above, when the idle device detects an idle channel during CCA execution time T _CCA , the first device 11 successfully transmits a packet. When the second device 12 performs CCA to transmit the generated packet, the second device 12 detects a busy channel due to the packet transmission of the first device 11 and increases the backoff index and repeats the random backoff. . The second device 12 discards and discards the generated packet if the CCA attempt is made with M + 1 failure. In the above, M has a value of 4 as a default value as the CCA performance maximum value (macMaxCSMABackoffs).

3 is a timing diagram of each device when operating in an unslotted CSMA / CA scheme in IEEE 802.15.4 according to an embodiment of the present invention.

When the channel becomes idle during CCA, the first device 11 immediately transmits the packet generated by itself. However, in the unslotted CSMA / CA, since the backoff slot unit time is not synchronized, as shown in FIG. 3, the other second device 12 or the third device 13 is connected to the first device 11. Attempting CCA during packet transmission may occur.

That is, if there are three devices 11, 12, and 13 as shown in FIG. 3, the first device 11 performs a CCA before transmitting the packet and transmits the packet when the idle channel is detected during the CCA execution time. do. At this time, when the second device 12 performs CCA, the busy device detects the busy channel and increases the backoff index to repeat the random backoff. In addition, when the third device 13 performs the CCA, the busy device detects a busy channel due to the ACK transmission of the first device. As a result, if the channel is idle during the CCA execution time, the packet transmission is always successful.

Because unslotted CSMA / CA is asynchronous with backoff slot time of all devices, the backoff slot time for one device is not synchronized with the backoff slot time of another device on the network. . The nonslot based CSMA / CA algorithm is as follows. The device generates a packet to send each time, waiting for a random number of backoff slots, called backoff counters, in the range of 0 to 2 ^BE -1, where BE represents the backoff exponent. The BE represents a backoff index and is initialized to macMinBE (default '3').

The backoff counter is decremented regardless of the channel state, and when the backoff counter reaches zero, the first device uses one CCA as shown in FIG. 3 to check whether the channel is busy or not. To perform.

Each device generates only one packet and attempts to store and send it. If the packet is not transmitted, the generated packet is lost. That is, each device loses a packet generated when M + 1 failures occur in performing CCA. M is the CCA performance maximum value (macMaxCSMABackoffs) and has a value of '4' as a default. Each packet is lost when it fails in M + 1 backoff processes (= M + 1 CCA attempts), so a packet loss probability (P _loss ) can be obtained.

Meanwhile, as described above, when the CCA execution time is designed to be longer than the switching time from data reception to ACK transmission in the unslotted CSMA / CA scheme in IEEE 802.15.4, the packet loss probability (P _loss) ) And transmission delay time, it is important to determine the optimal number of devices in the WPAN network configuration and to predict performance measures such as battery life. That is, when the CCA execution time is designed to be longer than the switching time from data reception to ACK transmission, the optimal WPAN network can be configured by predicting the optimal number of devices and battery life. This optimal number of device and battery life prediction algorithms will be described with the flowchart of FIG.

4 is a diagram illustrating an optimal number of devices and a battery life in a network when a CCA execution time is designed to be longer than a switching time from data reception to ACK transmission in an unslotted CSMA / CA scheme according to an embodiment of the present invention. It is a flowchart showing the process of calculating the performance measure such as.

Prior to the description of the process of calculating the performance measures such as the optimum number of devices and battery life prediction, the following matters are assumed. i) n sensor device is associated with a PAN coordinator, and ii) at irregular or continuous arrival of a packet with the same information, each device generates a packet with a Poisson process following the λ rate. Iii) the data packet size is invariant so that the transmission time of the data packet is fixed at T _tx , and iv) the switching time of the PAN coordinator from the received data to the transmitted data is ignored.

Under the above assumption, the packet generation process S41 of each device is generated through a Poisson process having a λ rate. That is, the probability process {X (t); If t≥0} and looking at the point of packet departure (packet loss beyond the completion or maximum backoff count) of the device, X (t) becomes a regeneration process and a regeneration cycle of the inter-departure time. (regenerative cycle). That is, {X (t); t≥t _{k + 1} } The process is {X (t); probable probabilistic replication of t ≧ _{t k} }, where {t (k); n = 1,2,3 ....} is the regeneration point. Thus {X (t); t≥0} is a regeneration process with regeneration cycles corresponding to packet inter-departure times.

[Equation 1]

X (t) = {Idle: When no device transmits at time t,

         Backoff: When the device does a backoff process at time t,

         CCA: When the device is in CCA state at time t,

         Transmission: when the device is transmitting a packet at time t}

The packet generated through the Poisson process having the λ rate is busy when the CCA is performed in a network structure in which the CCA execution time of the device is longer than the switching time which is a time from data reception to ACK transmission. Packet transmission occurs when an idle state is detected that is not busy.

After the packet generation (S41) is made, the estimated delay of the packet transmission success probability α, the backoff time period after performing the CCA used to calculate a performance measure including the optimal number of devices, battery life prediction, etc. E [D], packet loss probability P _loss is calculated (S42). In particular, in calculating the performance measure including the optimal number of devices and battery life prediction, the probability of packet transmission success α, which is the probability of occupying a channel during CCA enforcement, is the most important factor. Since all devices have the same probability of transmission, the probability α of success in CCA enforcement is expressed as shown in Equation 2 below.

[Equation 2]

In the above, λ is the rate used in the Poisson process at the time of packet generation, T _CCA is the CCA execution time, T _TX is the packet transmission time, T _ACK is the ACK transmission time, P _loss is the packet loss probability, E [D] is A backoff time period, which represents the expected delay from the arrival point to the device just before the packet is sent or lost.

As shown in FIG. 5, when there is a first device 11 that performs CCA, transmits data, and transmits ACK, the other devices 12 and 13 perform a CCA at a time T _CCA and a data transmission time T _TX ) and channel busy during ACK transmission time (T _ACK ). Since every device has an equal opportunity to transmit during the regeneration cycle of the tagged device, every other n-1 device will be lost to one regeneration point (P _loss , the probability of packet loss, Will be transmitted successfully with a probability of P _loss ). Therefore, the successful average number will be (n-1) (1-P _loss ), so the channel occupancy time by n-1 devices is (n-1) (1-P _loss ) (T _CCA + T _TX + T _ACK ) to obtain an equation such as [Equation 2].

The equation for calculating E [D], which is the expected delay, is shown in Equation 3 below.

[Equation 3]

In the above, Wj = min {2 ^j W _min , W _max } and W _min = 2 ^BEmin , W _max = 2 ^BEmax . BE _min and BE _max mean a minimum backoff index macMinBE and a maximum backoff index macMaxBE, respectively. Their default values have 3 and 5 respectively.

In addition, sigma represents the length of a backoff slot. The first summation of Equation 3 corresponds to the situation in which the packet is successfully transmitted, and the second summation of Equation 3 is packet lost after M + 1 attempts of CCA failure. It fits the situation.

As described above, each packet is lost when it fails in the M + 1 backoff process (= M + 1 CCA attempts), and thus the packet loss probability (P _loss ) is obtained as shown in Equation 4 below.

[Equation 4]

[Equation 2], [Equation 3] and [Equation 4], [alpha] in [Equation 2] is expressed as a factor item of E [D] and P _loss . In addition, P _loss of E [D] and [Equation 4] of [Equation 3] is expressed as a factor item of α. Therefore, by iterating the nonlinear formulas [Formula 2], [Formula 3] and [Formula 4], we can calculate all the necessary values α. In addition, the expected delay E [D] and the packet loss probability P _loss can be obtained.

As a result, when the equations [Equation 2], [Equation 3] and [Equation 4] which are nonlinear equations are iterated, numerical results as shown in FIGS. 5 (b) and 5 (c) are obtained. In consideration of all of these results, an optimal number of devices satisfying the limitations of a specific environment is calculated (S43). That is, when generating a packet through a Poisson process having a λ rate, each packet transmission probability α according to the number of devices for each λ rate, an expected delay E [D], and a packet loss By calculating the probability P _loss , the relationship as shown in FIG. 5 can be known, and the number of devices meeting the network configuration environment is selected (S43). The optimal network can be configured using the calculated number of devices (S43).

In addition to calculating the number of devices, the present invention may predict battery life of each of the calculated number of devices using the parameter values. To this end, the amount of energy consumed per unit time in the device is calculated (S44), and the battery life of the device is predicted using the calculated amount (S44).

The calculation of the amount of energy consumed per unit time (S44) includes the energy consumption per unit time (ETX) at the time of transmission, the energy consumption per unit time (ERX) at the time of reception, the energy consumption per unit time (ECC) at the time of performing CCA, children The energy consumption per unit time in one device is calculated in consideration of the energy consumption per unit time in the idle state and each operation time.

That is, when the number of devices is determined, the amount of energy consumed per unit time is calculated. Since energy consumption is the most important measure of performance, it is the average energy consumption per millisecond. Here, unit time was made in milliseconds.

E _TX , E _RX , E _CCA , and E _idle are the energy consumption per millisecond required to transmit (TX), receive (RX), CCA, and idle. The E _idle is consumed during the backoff process per millisecond.

Therefore, the amount of energy consumed per unit time (E) is calculated (S44) as shown in [Equation 5].

[Equation 5]

The N ^CCA is an average number of CCAs until successful transmission or loss, as shown in Equation 6 below.

[Equation 6]

Using the amount of energy (E) consumed per unit time calculated by Equation 5, L ^battery which is a life expectancy time of the ^battery may be obtained by Equation 7 below (S45).

[Equation 7]

In this case, E ^battery is a battery energy amount.

As a result, when the CCA execution time is designed to be longer than the switching time from data reception to ACK transmission in order to protect the ACK in the unslotted CSMA / CA scheme in IEEE 802.15.4, Battery life can be calculated | required by said Formula (7).

On the other hand, in order to prove the above description, it will be described by comparing the calculation result and the simulation result. 5 is a graph illustrating a result of a calculation and a result of directly simulating the calculation, according to an embodiment of the present invention.

Referring to the graph of FIG. 5, as the default parameter values in the 2.4 GHz frequency channel, we use macMinBE, macMaxBE, macaxCSMABackoff, and α as 3, 5, 4, and 0.32 (ms), respectively. The parameters T _CCA , T _TX , and T _ACK use 0.25 ms, 1.12 ms, and 0.352 ms, respectively. 5 shows the results when λ = 0.01, λ = 0.02, and λ = 0.03 per millisecond.

The fraction of time that a channel transmits a data load, defined as normal system throughput S. N (1-P _loss ) packets are transmitted on average during one regeneration cycle. Therefore, the normal system throughput S can be obtained by the following [Equation 8].

[Equation 8]

For all performance measures, it can be seen that the calculation results closely match the simulation results. As the number of nodes and the packet reception rate increase, as we expected, the normal system throughput S, the expected delay E [D], and the packet loss probability P _loss increase.

Since there are not many packet arrivals, it can be seen that the energy consumption does not depend on the number of devices when [lambda] = 0.01 and [lambda] = 0.02. As the number of devices increases to λ = 0.03, more devices find more channel busy and perform more random backoffs to reduce energy consumption. 5 (e) shows the life time L ^battery as the number of devices increases when λ = 0.01, λ = 0.02, λ = 0.03 and the battery capacity is 560mAh at 3.0V.

E _TX , E _RX , E _CCC , and E _idle are 0.0100224mJ, 0.0113472mJ, 0.0113472mJ, and 0.000057637mJ. This shows that battery life is more dependent on the packet arrival rate than on the number of devices. This is the reason why at least energy consumption is needed during the backoff process, so energy consumption is less affected by increased delay. In conclusion, it is useful to determine the optimal number of devices that can be met by the system supporting QoS required in the expected delay (E [D]) and packet loss probability (P _loss ). For example, if P _loss ≤ 2% and E [D] ≤ 10 ms are required when λ = 0.02 per millisecond, the optimal number of devices on the network is 16 from FIGS. 5 (b) and 5 (c). Is determined. Through this case, it can be seen that the average energy consumption E ^slot per backoff ^slot is 1.18 * 10 ⁻⁴ mJ / slot from 5 (d) and the battery life is 60 days from FIG. 5 (e).

In the above description of the present invention, specific embodiments have been described, but various modifications may be made without departing from the scope of the present invention. Therefore, the scope of the present invention is not to be determined by the embodiments described above, but will be apparent in the claims as well as equivalent scope.

As described above, the present invention provides the ACK protection by designing a CCA execution time longer than the switching time from data reception to ACK transmission in a non-slot-based (non-beacon mode) CSMA / CA of IEEE 802.15.4. There is an effect that can prevent the collision between each device. In addition, it is possible to calculate the optimal number of devices under a specific network environment, and by predicting the battery life of such a device, there is an effect that can build an optimal network environment.

Claims (10)

In the method for protecting the ACK of each device in the network configuration by the unslotted CSMA / CA scheme of IEEE 802.15.4, The CCA execution time of the device is designed to be longer than the switching time, which is the time from the data reception to the ACK transmission to prevent the packet collision between each device, characterized in that the ACK protection method. The method of claim 1, wherein the device discards the generated packet when the number of CCA executions to be failed due to channel busy becomes the number of 'CCA execution maximum value + 1'. In the method of predicting the performance measure of each device in the network configuration by the unslotted CSMA / CA scheme of IEEE 802.15.4, A first process in which a packet used under a structure in which a CCA execution time of the device is designed to be longer than a switching time from a data reception to an ACK transmission is generated through a Poisson process having a λ rate; A second step of calculating a packet transmission probability α, an expected delay E [D] having a backoff time period, and a packet loss probability P _loss after the device performs CCA; A third process of calculating the number of devices corresponding to the network configuration environment in consideration of the packet transmission probability α, the expected delay E [D], and the packet loss probability P _loss Performance measure prediction method comprising a. 4. The method of claim 3, wherein T _CCA is CCA execution time, T _TX is packet transmission time, T _ACK is ACK transmission time, P _loss is packet loss probability, and E [D] is a backoff time period. When the delay (delay), the packet transmission success probability α,

Performance scale prediction method calculated by 4. The method of claim 3, wherein M is the maximum backoff performance (4 as the default), T _CCA is the CCA execution time, sigma represents the length of the backoff slot, and Wj represents min {2 ^j W _min , W _max }. In this case, when W _min = 2 ^BEmin and W _max = 2 ^BEmax , and BE _min and BE _max represent the minimum backoff index macMinBE and the maximum backoff index macMaxBE, respectively, the expected delay E [D] is

Performance scale prediction method calculated by 4. The packet loss probability P _loss of claim 3, wherein when M is a backoff performance maximum value (4 as a default), when a CCA attempt fails at a point M + 1, Is,

Performance scale prediction method calculated by The probability of each packet transmission success α according to the number of devices for each λ rate when generating a packet through a Poisson process having a λ rate. Predictive Delay E [D], Packet Loss Probability P _loss , and then select the number of devices that meet the network configuration. The method of claim 3, wherein the performance scale prediction method is as follows. Energy consumption per unit time (E _TX ) when transmitting, energy consumption per unit time (E _RX ) when receiving, energy consumption per unit time (E _CCA ) when performing CCA, energy consumption per unit time when idle A fourth process of calculating energy consumption per unit time in one device in consideration of E _idle and each operating time; A fifth process of calculating a battery life time by dividing an energy amount of a battery of the device by the energy consumption per unit time Performance measure prediction method further comprising. The method of claim 8, wherein E _TX , E _RX , E _CCA , and E _idle are energy consumption per unit time (per millisecond) required when transmit (TX), receive (RX), CCA, and idle are performed. _CCA is the CCA execution time, T _TX is the packet transmission time, T _ACK is the ACK transmission time, E [D] is the expected delay, which is the backoff time period, and N ^CCA may be a successful transmission or loss. When indicating the average number of CCA until, the energy consumption per unit time,

A method of predicting performance scale calculated by. 10. The method of claim 9, wherein when M is the backoff performance maximum value (4 as the default), when the CCA attempt fails at the M + 1 point, the N ^CCA is

A method of predicting performance scale calculated by.

Images