CN115226138A - CoAP-oriented efficient data collection method for Internet of things - Google Patents

Abstract

The invention discloses a CoAP-oriented high-efficiency data collection method for the Internet of Things. The method adopts NB-IoT as a low-power-consumption network communication link, and the upper layer adopts a CoAP application layer communication protocol, comprising the following steps: Step 1: CoAPPDU filtering; Step 2: Combine and compress; Step 3: Adaptive upload. The method filters similar data of the Internet of Things, reduces redundant information, and solves the problems of low payload rate and large header overhead of lightweight network communication protocols in traditional low-power networks. Reduce the amount of data transmitted over the network, and make a good balance between uplink data delay and device power consumption.

Description

A CoAP-Oriented Efficient Data Collection Method for Internet of Things

technical field

The invention belongs to the technical field of the Internet of Things, and relates to a data collection method of the Internet of Things, in particular to a CoAP-oriented efficient data collection method of the Internet of Things.

Background technique

With the rapid development of IoT technology, there are tens of billions of lightweight resource-constrained terminal devices in the real world, such as sensors and electricity meters. Such devices put forward higher requirements for network communication energy consumption, delay, and data transmission efficiency. Low-power wide-area network (LPWAN) technology is widely used in the IoT field due to its low energy consumption, wide coverage, and large capacity. Currently, LPWAN includes NB-IoT, eMTC, etc. deployed in licensed frequency bands, and LoRa deployed in unlicensed frequency bands. , Sigfox, etc. NB-IoT is a low-power wireless wide-area network technology developed based on existing LTE specifications and facilities. It has the advantages of wide coverage, large connection, low power consumption, and low cost, and is well suited for various lightweight resource constraints. IoT end devices are required. At the same time, resource-constrained lightweight IoT devices also put forward higher requirements for application layer network protocols, and the CoAP protocol for resource-constrained devices is widely used. CoAP runs on the UDP protocol, the message format is compact, and it is designed based on the REST architecture, which can meet the basic functions of IoT data upload, configuration and delivery.

Because the IoT data is generally short and the information level is low, the IoT lightweight devices have the problem of low data transmission payload rate. At the same time, LPWAN puts forward higher requirements for energy consumption, while traditional low-power network communication links and upper-layer IoT application layer protocols generally lack data filtering and adaptive data upload mechanisms. Excessive data collection frequency configuration leads to redundant information of IoT devices, causing network congestion and shortening the service life of low-power devices. Too low frequency configuration leads to a sharp increase in the uplink delay of the device and reduces the real-time performance of data. Furthermore, the LPWAN bandwidth is limited, which requires the packet load. Currently, SenML and CBOR IoT data compression technologies are widely used. SenML is a simple information model for sensor data and device parameter transmission. CBOR (Concise Binary) Object display) is a binary-based data compression algorithm, but the above method has poor effect on the compression of short and small data of the Internet of Things, and does not support the function of combining and compressing multiple data packets. Therefore, a more efficient IoT data transmission method is required.

SUMMARY OF THE INVENTION

Aiming at the low data transmission effective load rate caused by short IoT data, and the lack of adaptive data collection and adaptive data uploading in traditional IoT lightweight communication protocols, the present invention provides a CoAP-oriented IoT efficient data collection method . The method filters similar data of the Internet of Things, reduces redundant information, and solves the problems of low payload rate and large header overhead of lightweight network communication protocols in traditional low-power networks. Reduce the amount of data transmitted over the network, and make a good balance between uplink data delay and device power consumption.

The purpose of this invention is to realize through the following technical solutions:

A CoAP-oriented IoT efficient data collection method adopts NB-IoT as a low-power network communication link, and the upper layer adopts CoAP application layer communication protocol, including the following steps:

Step 1: CoAP PDU Filtering

Compare the data content of the CoAP protocol data unit PDU to be sent by the lightweight terminal device with the data sent last time. If the relative error range is less than rate% or the absolute difference is less than diff, the data will not be further processed and will wait for new data. The data arrives, otherwise go to step 2;

Step 2: Merge Compression

Step 2.1: If the time interval from the last data transmission is greater than MAX_Interval, go to Step 2.4, otherwise, judge whether the number in the CoAP PDU queue exceeds the MAX_PDU_COUNT threshold, if it exceeds the threshold, go to Step 2.3, otherwise go to Step 1;

Step 2.2: Extract key information of all PDUs in the CoAP PDU queue, such as event unique identifier token, data packet sequence number mid, timestamp, etc.;

Step 2.3: If the time interval from the last data transmission is less than MIN_Interval, go to step 1;

Step 2.4: Encapsulate all the key information of CoAPPDU into a single PDU to form PDU Merged;

Step 2.5: Use the SenML compression algorithm to reduce the timestamp load overhead in the payload in the PDU Merged, and use the CBOR compression algorithm to reduce the overall load overhead, and finally form PDU Compressed;

Step 3: Adaptive upload

Step 3.1: If the PDU Compressed arrival time is greater than MAX_Interval from the last data sending interval, increase MAX_Interval to twice itself. If the PDU Compressed arrival time is less than MAX_Interval from the last data sending interval, adjust MAX_Interval to the initial minimum value ;

Step 3.2: Send the PDU Compressed to the target server.

Compared with the prior art, the present invention has the following advantages:

1. By using the method of the present invention, the effective load rate of low-power network transmission can be greatly improved, and the header overhead of the transmission layer and the link layer can be reduced.

2. By using the method of the present invention, similar data of the Internet of Things can be filtered, the data upload period can be adjusted adaptively, a better balance can be achieved between data delay and power consumption, and the goal of efficient data transmission can be achieved.

Description of drawings

1 is a flowchart of an adaptive data collection method based on filtering and compression;

Fig. 2 is the data filtering effect comparison diagram of the adaptive data collection method based on filtering and compression;

Fig. 3 is a network performance comparison diagram of an adaptive data collection method based on filtering and compression;

Figure 4 is a comparison diagram of energy consumption and delay of adaptive data collection methods based on filtering and compression.

Detailed ways

The technical solutions of the present invention will be further described below in conjunction with the accompanying drawings, but are not limited thereto. Any modification or equivalent replacement of the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention shall be included in the present invention. within the scope of protection.

The present invention provides a high-efficiency data collection method for CoAP packet data unit (PDU) - an adaptive data collection method based on filtering and compression (FC-ADC, Adaptive Data Collection algorithm based on Filtering and Compression), using NB-IoT as the In the low-power network communication link, the upper layer adopts the CoAP application layer communication protocol, as shown in Figure 1, the method includes the following steps:

Step 1: CoAP PDU filtering.

Step 1.1: According to the specific needs of the IoT business, adaptively adjust the rate and diff parameters. The present invention provides a specific implementation. When it is desired to reduce the network data traffic to the greatest extent, the rate needs to be configured to 40%, and the diff needs to be configured to 20. When the network data traffic is expected to be appropriately reduced, the rate needs to be configured not lower than 20% parameter, diff is configured to 100.

Step 1.2: When an IoT uplink data is generated, the adaptive data collection method based on filtering and compression is called for processing.

Step 1.3: Determine whether the data needs to be uploaded according to the rate and diff parameters, if necessary, go to Step 2, otherwise return to Step 1.2.

Step 2: Merge Compression.

Step 2.1: Initialize MAX_Interval to 10 seconds, MIN_Interval to 1 second, and MAX_PDU_COUNT to be 4.

Step 2.2: If the time interval from the last data transmission is greater than MAX_Interval, go to Step 2.5.

Step 2.3: When the time interval from the last data transmission is less than MAX_Interval, determine whether the number of CoAP PDU queues exceeds the MAX_PDU_COUNT threshold. If it exceeds the threshold, go to step 2.4, otherwise continue to wait for the next CoAP PDU to arrive.

Step 2.4: When the time interval from the last data transmission is less than MIN_Interval, return to step 1.

Step 2.5: Extract all PDU data arrival time, data content, data packet serial number mid and data packet identification token information in the CoAP PDU queue, and encapsulate it into a Json text.

Step 2.6: Use Json content as PDU Merged payload information.

Step 2.7: Use SenML (RFC 8428) compression algorithm to reduce timestamp load overhead in the payload in PDU Merged, and use CBOR (RFC 8949) compression algorithm to reduce overall load overhead, and finally form PDU Compressed.

Step 3: Adaptive upload.

Step 3.1: If the PDU Compressed arrival time is greater than MAX_Interval from the last data transmission time interval, increase MAX_Interval to twice itself. If the PDU Compressed arrival time is less than MAX_Interval from the last data transmission interval, adjust MAX_Interval to the initial minimum value.

Step 3.2: Send the PDU Compressed to the target server.

Example:

In this embodiment, a low-power NB-IoT link is used for the data link layer, UDP and TCP are used for the transport layer, and CoAP is used for the application layer protocol. Moderate traffic (10 tasks) and high traffic (30 tasks) scenarios are set. Three relative error values of 0.3, 0.4, and 0.6 are set, and the terminal device adopts the REFIT public data set. In the data filtering stage, compared with the AMID algorithm based on the pull method under three configurations of relative error values, the AMID algorithm and the CBOR algorithm reduce the data volume by 99.57% and 99.85%, respectively, and the root mean square error with the original data is 283.3 , 93.67, this filtering algorithm is more accurate than the AMID filtering algorithm, and the amount of data is smaller.

In the stage of data compression and merging and self-adaptive uploading, five sets of comparative experiments are set up, as shown in Figure 3, from left to right: CoAP/UDP medium and high traffic scenarios do not use the method of the present invention, CoAP/TCP medium, Performance comparison diagram of header overhead, number of network data packets, compression rate and uplink data delay when CoAP/TCP adopts the method of the present invention in a high-traffic scenario without using the method of the present invention. Figure 4 depicts the comparison between the average data delay and the energy consumption of the NB-IoT communication module in different data upload cycles. From the figure, we can see that the adaptive data collection algorithm based on cross-layer perception can make a good trade-off between energy consumption and data delay. Although a lower data upload cycle can reduce energy consumption, it will lead to excessive Data delay, higher data collection frequency requires NB-IoT to be in the RRC connection state more, resulting in greater energy consumption.

The invention reduces the extra overhead of the network transmission header by 70.85%, reduces the number of data packets in the network by 99.85%, and the data compression rate reaches 40.5%. Compared with the network high-frequency data transmission mode, the method of the invention can effectively increase the equipment sleep time of the NB-IoT communication module by 50.8%, and can effectively reduce the uplink data delay by 41.1% compared with the low-frequency data transmission mode, thereby ensuring the efficient transmission of Internet of Things data.

Claims (4)

1. a CoAP-oriented Internet of Things efficient data collection method is characterized in that the method comprises the steps: Step 1: CoAPPDU filtering Compare the data content of the CoAP protocol data unit PDU to be sent by the lightweight terminal device with the data sent last time. If the relative error range is less than rate% or the absolute difference is less than diff, the data will not be further processed and will wait for new data. The data arrives, otherwise go to step 2; Step 2: Merge Compression Step 2.1: If the time interval from the last data transmission is greater than MAX_Interval, go to Step 2.4, otherwise, judge whether the number in the CoAPPDU queue exceeds the MAX_PDU_COUNT threshold, if it exceeds the threshold, go to Step 2.3, otherwise go to Step 1; Step 2.2: Extract the key information of all PDUs in the CoAPPDU queue; Step 2.3: If the time interval from the last data transmission is less than MIN_Interval, go to step 1; Step 2.4: Encapsulate all the key information of CoAPPDU into a single PDU to form PDUMerged; Step 2.5: Use the SenML compression algorithm to reduce the timestamp load overhead for the payload in the PDUMerged, and use the CBOR compression algorithm to reduce the overall load overhead, and finally form PDUCompressed; Step 3: Adaptive upload Step 3.1: If the PDUCompressed arrival time is greater than MAX_Interval from the last data sending time interval, increase MAX_Interval to twice itself; if the PDUCompressed arrival time is less than MAX_Interval from the last data sending time interval, adjust MAX_Interval to the initial minimum value; Step 3.2: Send PDUCompressed to the target server. 2. The CoAP-oriented Internet of Things efficient data collection method according to claim 1, characterized in that in the step 1, when it is expected to reduce network data traffic to the greatest extent, the rate is configured to be 40%, and the diff is configured to be 20; Expect to appropriately reduce network data traffic, configure rate to be no less than 20% parameter, and configure diff to 100. 3. the CoAP-oriented Internet of Things efficient data collection method according to claim 1, is characterized in that in described step 2, PDU key information comprises one or more in event unique identification token, data packet sequence number mid, timestamp. kind. 4. The CoAP-oriented Internet of Things efficient data collection method according to claim 1, characterized in that the method adopts NB-IoT as a low-power network communication link, and the upper layer adopts CoAP application layer communication protocol.

Images