TWI569656B - Estimate the number of network devices - Google Patents

Description

Method of estimating the number of network devices

The present invention relates to a method of estimating the number of network devices, and more particularly to a method for actively estimating how many devices currently want to access a network, thereby making the congestion control more accurate and efficient.

At present, the design of wireless networks has not actively estimated how many devices want to access the network, so it has the following two major drawbacks: (1) It is impossible to know whether the current radio resources can cope with the connection. The number of devices; (2) the efficiency of congestion control is low.

The existing random access method cannot know the number of devices that need to be connected. Therefore, when congestion occurs, the mechanism of congestion control cannot be accurately regulated, and it is often possible to overestimate the congestion situation, resulting in low efficiency of random access resources. It takes a long time to process a large number of random access requirements. On the other hand, the prior art may also be due to underestimation of the congestion situation, resulting in no effective solution to the congestion control. The root cause of these shortcomings is because they don't know how many devices they want to access the network.

In view of the above-mentioned problems of the prior art, the object of the present invention is A method for estimating the number of network devices is provided, so that the subsequent congestion control can adjust the parameter settings more accurately and effectively, thereby effectively solving the congestion situation.

The method for estimating the number of network devices of the present invention is applicable to a base The station includes the following steps: sending a random access signaling transmission probability and a random access resource block; after the network device generates the random variable by itself, determining whether to send the random access preamble signaling according to the random access signaling transmission probability The base station calculates the number of network devices currently connected according to the number of successful random access resources, the number of idle random access resources, and the number of collision random access resources. The base station can choose to perform multiple rounds of estimation, and each round can gradually obtain accurate estimation results by adjusting the random access signaling transmission probability.

In view of the above, the method of estimating the number of network devices according to the present invention may have one or more of the following advantages:

1. The present invention can estimate a relatively large number of devices with a relatively small portion of random access resources. At the same time, since the estimation error mechanism of the present invention is only related to how much resources are used for estimation, regardless of the actual number of devices, the present invention can be applied to an extremely large number of applications of Internet of Things or machine type communication.

2. The present invention directly estimates the connection demand, regardless of the amount of connection demand and the speed of change, the base station can estimate the number, so the random access rules can be adjusted according to the base (such as delay connection time, retransmission interval) Or the number of exclusive random access resources) to avoid the transmission of random access preamble signaling. Plugging, or random access resources are inefficient.

3. The base station of the present invention can also estimate the number of emergency or high priority connections, and give more random access resources for connection, so that an urgent or high priority machine device will not be blocked due to network congestion. Unable to connect quickly.

FIG. 1 is a schematic diagram of a random access mechanism of the long term evolution technology.

2 is a schematic diagram of signaling exchange of a method for estimating the number of network devices according to the present invention.

3 is a flow chart of a random access number estimation mechanism of the method for estimating the number of network devices according to the present invention.

4 is a schematic diagram showing the simulation result of the estimation error of the method for estimating the number of network devices of the present invention.

This technical description is based on the random access of 4G system Long Term Evolution (LTE). However, it is noted that the present invention can be applied to any random access system, such as random 2G, 3G communication systems. Access mechanism, with traditional IEEE 802.11 (WiFi) systems.

4G system Long Term Evolution (LTE) random access mechanism (RACH procedure), can be subdivided into four signaling See step 1 for the replacement procedure. The UE (user equipment) is an abbreviation of the user device, and refers to all possible terminal devices, such as a machine-to-machine device or a machine type communication device (MTC device). Wait. The eNB is the English name of the base station in the Long Term Evolution system.

As shown in Figure 1, the existing LTE random access mechanism steps are as follows:

The first step is random access preamble signaling (RACH preamble), which is sent to the base station when the mobile phone or device wants to connect to the network; in the same base station, the random access preamble With only 64 options, when there are many devices in a base station range, it is possible for several devices to select the same random access preamble signaling.

The second step is a second message (Msg2). When the base station detects that some preamble signaling is transmitted, the base station broadcasts a second message to the (some) transmitted detected preamble signaling. The mobile phone or device, please use the upload resource allocated in the second message to upload the next third message (Msg3); and the device using the same random access preamble signaling will get the same message.

The third step is a third message (Msg3). After receiving the correct second message, the device transmits a third message to the base station, and attaches its own (temporary) identity to make a connection request.

In the fourth step, the base station replies with the fourth message to the mobile phone or device that transmitted the third message to confirm whether the connection is successful. If there are several devices used For the same random access preamble signaling, the base station will only reply to the success message of the device that can successfully judge the TC-RNTI, otherwise the UE will fail to compete.

After the above four steps, if the user device can be successfully connected, it will change from the idle state to the connected state.

As described above, in the first step, the device randomly selects one of the 64 preamble signalings that are shared, so that when the number of devices is large, the same preamble signaling may be used by two. The above device is selected for transmission, which we call a preamble collision. The phenomenon of collision of the preamble signaling may cause the base station to fail to correctly decode the preamble signaling of the collision, and the random access preamble signaling transmission fails. Therefore, if a large number of devices, such as Machine-to-Machine Communications (M2M) or Internet of Things (IoT) devices, simultaneous network access and simultaneous random access preambles The transmission will cause the collision of the preamble signaling to be very serious, not only causing the random access card of the machine device to fail in the transmission of the preamble signaling, but also causing the failure of the connection of the general mobile phone (Human-to-Human Communications, H2H).

Past research and communication standards have noticed this problem, so some methods have been adopted to avoid congestion. The main practice is to delay the time that the machine device transmits the preamble signaling, such as delay to non-peak time. (2) Extend the interval between retransmissions of the machine to avoid causing more serious congestion caused by a large number of retransmissions in a short period of time, and (3) assigning different access resources to H2H and M2M communication to avoid the access of the machine to affect normal human communication. However, because past mechanisms cannot know how many devices want random access, base stations may overestimate or underestimate congestion. If the congestion situation is underestimated, the base station may set the delay time, interval time, or exclusive random access resources of the signaling to be too small, so the congestion situation cannot be effectively solved. Conversely, if the congestion situation is overestimated, the base station may Setting these three parameters too large results in a large amount of idleness and inefficiency of random storage and resources.

The invention can estimate the number of devices to be accessed by the network, so that The previous mechanism can adjust the parameter setting of the congestion control more accurately and effectively, for example, setting the appropriate delay time, interval time, and exclusive resources to solve the congestion situation and improve the efficiency of random access resources. The signaling exchange of the present invention is shown in FIG. 2, and the flowchart is shown in FIG. 3. The detailed steps are as follows: The base station starts the estimation method, and places relevant estimation information in the system information block, and the information includes the following contents: The random access resource block range R used to estimate the number of devices, and the random access signaling transmission rate P. The random access resource is a two-dimensional plane, which are respectively available sequences of time and random access preamble signaling.

When the device receives the estimated information broadcasted by the base station, the following procedure is performed: the device generates a uniformly distributed random variable v between [0, 1]. If the random variable v is less than the probability P, the device randomly selects a random access resource from the base station to broadcast the random access resource block used to estimate the number of devices, and performs random access preamble signaling to participate in the present Estimation mechanism; If the random variable v is greater than or equal to the probability P, the device does not transmit the random access preamble signaling temporarily, that is, does not participate in the estimation mechanism.

The base station counts it according to the random access preamble signaling transmitted by the device. The number of successful random access resources S, the number of idle random access resources, and the number C of collided random access resources in the random access resource block used to estimate the number of devices are broadcast. The successful random access resource indicates that exactly one device selects this resource for the transmission of random access preamble signaling. Idle random access resources represent random access resources that are not selected by any device. The collision random access resource indicates that more than two devices select this resource for random access preamble signaling.

The base station further makes the following judgment: if all random access resources collide (S=0, E=0, C=R), R is adjusted to be large or P is turned down, and [0019] is returned. If all random access resources are idle (S=0, E=R, C=0), if P is small, P is increased, back to [0019], if P is large enough (for example, P 0.2), you can guess This estimate is over.

If neither of the above occurs, proceed to [0024].

The base station estimates according to the following method: the base station assumes that M' devices participate in the estimation mechanism, and each device transmits from a random random access resource uniformly among R random access resources, then exactly S The preamble signaling is successful, and the probability of E preamble signaling being idle is: With the base station S, E, and the value of R, that it is possible for all of M 'number of the above calculation, from all the M' selected maximum P, (S, E | M ', R) of M' As an estimate Guess The number of devices participate in this estimation mechanism: Suppose there are actually M devices that want to transmit preamble signaling, but there are M ₁ devices that actually participate in the estimation mechanism. Because the probability of each device transmitting preamble signaling is P, the expected value of M ₁ is actually MP , so we directly estimate M is ,among them Is the estimated value of M.

The base station decides whether the estimated demand has been met. If it is desired to obtain a more accurate estimate, the base station can choose several estimation mechanisms to determine its estimated value. Estimated error About , where R _sum is when P The sum of the resources used to estimate the amount of access, for example, in P It took 10 rounds to estimate 40 random access resources per round, and the error was about 5%. This method can be used to speculate on how accurate the current estimate is and whether it has met the demand. If you want a more accurate estimate, you can set P = Go back and make another round of estimates.

The results of each estimate are averaged or calculated, and the estimate ends. If the total is estimated n times, the estimated value is =1 , ... , n, the number of successful and idle resources obtained in the jth time are S _j and E _{j respectively} , then it can be estimated as follows: Take the average of multiple rounds: Or take the maximum number of rounds:

The present invention is a mechanism design for estimating the random access demand, and its features and architecture can be summarized as follows:

1. The base station utilizes some of the existing random access resources for the connection device to perform the estimation mechanism test. The base station places test related information in the system information block, including the random access resource block R for estimation, and the probability P of the device transmitting random access signaling.

2. After the test, the base station observes the collision situation of the estimated random access resources, and calculates the number of successful, collided, and idle random access preamble resources.

3. Based on the number of all random access resources (R), the number of successful resources (S), and the number of idle resources (E), the base station uses the mathematical method described above to infer how many devices currently want to perform the network. Pick up (M).

[Examples]

The following describes an embodiment of the present invention by taking the LTE communication system as a background and the application of the M2M as an example.

The method of the present method in conjunction with the congestion control is as follows: Before the congestion control is started, the device is connected according to the random access protocol established by the foregoing existing LTE system.

When the base station detects the occurrence of congestion, the estimation method is started, and the random access of the machine device is prohibited. Base station broadcast is used to estimate The number of random access resource blocks of the device, and the probability P of the device transmitting random access signaling. The base station can set different random access resource block ranges and devices to transmit random access signaling for different types of devices according to different priorities of the devices.

After receiving the broadcast information of the base station, the device according to its own The category of random access resource blocks in this category is randomly selected to transmit random access preamble signaling.

According to the situation of the collision, the base station can be estimated according to the aforementioned mathematical method. The approximate number of devices in each category is determined, and the subsequent setting of random access parameters is determined. After obtaining an estimate of the amount of demand received, the congestion control method can be simply designed. If the number of demand is large, in order to avoid congestion, the base station should set a relatively large backoff window size or takeover probability (access opportunity for each device in the next random access preamble signaling). How many chances can be used to transmit random access preamble signaling); conversely, if the demand is not large or very small, the base station can set a smaller retransmission interval size or pick up probability to reduce the device retransmission. The time to wait.

The performance of the present invention is shown in Figure 4. At the beginning, there are 10,000 devices that want to be connected, and 54 pilot signals are estimated to be used each time. The length of the error bars is the standard deviation. It can be found that when the estimate of 10 rounds is made, 540 resources are used, so the theoretical error can be approximated as = 0.043, the simulation result of Fig. 4 can indicate that the estimation method of the present invention makes the error quite close to the theoretical value.

From the above examples, the advantages of the present invention can be found as follows: It can be seen from the experimental data that the present invention can be used with relatively few random parts. Access resources to estimate the relative number of devices. At the same time, since the estimation error of the mechanism of the present invention is only related to the estimation of how many resources are used, regardless of the actual number of devices, the present invention can be applied to an extremely large number of applications of Internet of Things or machine type communication.

The invention directly estimates the connection demand, no matter how much the connection demand is Widowed and variable speed, the base station can estimate the number, so the random access rules (such as delay connection time, retransmission interval, or the number of exclusive random access resources) can be adjusted to avoid random access. The transmission of preamble signaling is blocked, or the efficiency of random access resources is low.

The base station can also estimate the number of urgent or high priority connections. In this case, more random access resources are given for connection, so that urgent or high-priority machines cannot be quickly connected due to network congestion.

In summary, the method for estimating the number of network devices of the present invention, It can actively estimate how many devices currently want to access the network, which makes the congestion control more accurate and effective. The base station of the present invention further utilizes some of the existing random access resources for the connection device to perform an estimation mechanism test, and then estimates the number of devices to be connected. Moreover, the mechanism only needs a very small amount of signaling exchange, so the base station only needs to broadcast a small amount of estimation information, which is suitable for various types of connection requirements and connection numbers.

The above is intended to be illustrative only and not limiting. Any equivalent modifications or alterations to the spirit and scope of the invention are intended to be included in the scope of the appended claims.

Claims (7)

A method for estimating the number of network devices, applicable to a base station, comprising the steps of: transmitting a random access signaling transmission probability and a random access resource block; generating at least one of the plurality of network devices a random variable, and determining whether to send a random access preamble signaling to the base station according to the random variable and the random access signaling transmission probability; according to a successful random access resource quantity, an idle random access resource quantity, and The number of random access resources is used to estimate the number of the network devices; and the random access signaling transmission rate is adjusted according to the theoretical error value for multi-round estimation. For example, the method for estimating the number of network devices in claim 1 further includes the following steps: loading an estimated information in a system information block of the base station, and the estimated information includes the random access resource block. . The method for estimating the number of network devices according to Item 2 of the patent application further includes the steps of: estimating the number of the random access resource blocks, the number of the successfully random access resources, and the number of the idle random access resources. The number of network devices. The method for estimating the number of network devices according to item 3 of the patent application further includes the steps of: generating a plurality of possible number of the network devices; and picking the largest one from the possible numbers. For the number of these network devices. For example, the method for estimating the number of network devices in claim 4 of the patent scope further includes the following steps: Generating an estimated error value based on the total number of random access resources used in the multiple rounds; and determining whether to re-evaluate the number of the network devices based on the estimated error value. The method for estimating the number of network devices according to claim 1 of the patent scope further includes the step of: adjusting a plurality of random access parameters according to the estimated number of the network devices. The method for estimating the number of network devices according to claim 6 of the patent scope, wherein the parameters include a transmission delay time of preamble signaling, a retransmission interval time of preamble signaling, a probability that each transmission opportunity can transmit preamble signaling, and A resource dedicated to a particular priority device.