Method for modulation and coding adaptation selection in mobile ad hoc networks based on separation of control and data planes

Abstract

The application discloses a modulation and coding adaptive selection method for a mobile ad hoc network based on separation of a control plane and a data plane. In the mobile ad hoc network based on separation of the control plane and the data plane, a carrier sense multiple access (CSMA) mechanism is adopted in a control time slot to compete for sending fused situation information of itself, so as to complete survey and perception of neighbor nodes. The situation information includes link transmission quality and node queue length. In the data plane, the CSMA mechanism is also adopted, but the situation information of the control plane needs to be utilized to form certain 'network order', so as to reduce the decline of throughput caused by error packets and collision. On the basis of a traditional rate adaptive algorithm based on a signal-to-noise ratio (SNR), the algorithm is improved and applied to a mobile ad hoc network device based on separation of the control plane and the data plane.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and specifically discloses an adaptive modulation and coding selection method for mobile ad hoc networks based on the separation of control plane and data plane. Background Technology

[0002] A mobile ad hoc network (MANET) is a multi-hop, temporary, autonomous system. This network employs packet switching mechanisms from computer networks, with each user terminal functioning as both a router and a host. As a distributed network, MANET is an autonomous, multi-hop network without fixed infrastructure. Its purpose is to provide inter-terminal communication when existing network infrastructure is unavailable or inconvenient to utilize. Nodes employ different modulation and coding schemes (MCS) to cope with varying channel conditions.

[0003] The core of broadband ad hoc networks (BANs) is to effectively address channel complexity and network topology dynamics while providing guaranteed quality of service. Specifically, this involves designing two operating modes through two mechanisms: beamforming and spatial diversity, separating the data plane and control plane. In traditional Ad Hoc networks, the data plane and control plane cannot be separated, failing to address the two challenges faced by mosaic warfare communication networks. With BAN technology, the data plane and control plane are separated, allowing a series of new key technologies, such as dynamic adaptive modulation and coding (MCC), beamforming algorithms, network situational awareness-based spatiotemporal-frequency resource scheduling, service-aware end-to-end routing mechanisms, and context-aware dynamic network topology reconstruction, to be applied to mosaic warfare communication network design. The IEEE 802.11 standard does not define the selection method between different modulation and coding schemes for Ad hoc modes; the adaptive algorithms for MCS are primarily defined by hardware manufacturers within the equipment. Currently, mainstream MCS adaptive algorithms can be broadly categorized into three types: The first type is based on sender-end determination using historical information, including algorithms like AARF, Minstrel, and Onoe. These algorithms utilize ACKs to statistically analyze correct / failed transmissions and adjust the MCS accordingly. A drawback of this type is the need to calculate packet reception accuracy over a large time scale, which may lead to difficulties in quickly switching between high and low rates in ad hoc scenarios with sudden changes in channel conditions. The second type is receiver-end determination based on feedback, including algorithms like RBAR, OAR, and AAR. These algorithms use RTS / CTS for feedback and require changes to the RTS / CTS frame structure. This is incompatible with most 802.11 devices. The third type is based on SNR prediction feedback at the receiving node, including algorithms like SGRA and CHARM. These algorithms predict the SNR at the receiving node by utilizing the channel reciprocity between the sending and receiving nodes. The problem is that with a large network coverage area and uneven channel quality, there will be asymmetry in the noise power of the transmitting and receiving nodes, resulting in asymmetry in SNR and prediction deviation.

[0004] This invention addresses the bursty, rapidly changing channel conditions of Ad hoc networks. Ad hoc networks are multi-hop, decentralized, self-organizing wireless networks, also known as multi-hop networks; the entire network lacks fixed infrastructure, and each… node All nodes are mobile and can dynamically maintain contact with other nodes in any way. In such a network, due to the limited wireless coverage of terminals, two user terminals that cannot communicate directly can use other nodes for packet forwarding. Each node is simultaneously a... routerFurthermore, leveraging the structural feature of separating the data plane and control plane in this novel Ad hoc device, a modulation-coding adaptive selection method for mobile ad hoc networks based on the separation of the control plane and data plane is disclosed. This method 1) can directly utilize data time slots to obtain the SNR data at the receiving end, avoiding the prediction bias introduced by traditional algorithms based on predicting the SNR at the receiving node; 2) can dynamically adjust the SNR threshold according to the frame transmission rate, exhibiting strong dynamism and overcoming the problem that statistical algorithms cannot cope with rapidly changing channel conditions; 3) is applicable to novel mobile ad hoc network devices based on the separation of the control plane and data plane, rationally utilizing control plane information to decide the transmission rate of data time slots, thereby improving system throughput. Summary of the Invention

[0005] Purpose of the Invention: To overcome the problems of 1) slow response of algorithms such as Minstrel and AARF to the bursty and rapidly changing channel conditions of Ad hoc networks; 2) failure to distinguish between channel packet loss and collision packet loss; 3) problems of algorithms such as RBAR and OAR changing the frame structure of RTS / CTS and being unsuitable for most 802.11 devices; 4) the bias caused by algorithms such as SGRA and CHARM using the predicted SNR at the receiving node as the basis for MCS selection; and to be applicable to the new architecture of separating the control plane and data plane of this device, this invention discloses an adaptive modulation and coding selection method for mobile ad hoc networks based on the separation of the control plane and data plane.

[0006] The technical solution of this invention is: a modulation and coding adaptive selection method for mobile ad hoc networks based on the separation of control plane and data plane. In a mobile ad hoc network (MANET) with separated control plane and data plane, a carrier sense multiple access (CSMA) mechanism is used in the control time slot to compete for the transmission of its fused situational information, completing the survey and perception of neighboring nodes. The situational information includes link transmission quality and node queue length. To ensure that each node has the opportunity to transmit situational information, the situational information must be short enough and real-time enough, and each node transmits it at most once in the control time frame, while also considering backoff. In the data plane, the CSMA mechanism is also used, but the situational information from the control plane needs to be utilized to form a certain "network order" and reduce the throughput drop caused by packet errors and collisions. Based on the traditional rate adaptive algorithm based on signal-to-noise ratio (SNR), the algorithm is improved and applied to mobile ad hoc network devices with separated control plane and data plane, including the following steps:

[0007] Step 1: Construct a mapping table within the mobile ad hoc network device that shows the current link SNR low threshold and the index values ​​of all supported modulation and coding schemes (MCS).

[0008] Step 2: Read the latest updated SNR from the transmitting node to the receiving node from the situation frame received in the control time slot;

[0009] Step 3: Traverse the mapping table established in Step 1 and select the MCS index value corresponding to the current SNR as the base rate.

[0010] Step 4, Sampling and Transmission Process: Transmit c0 / c1 / c2 / c3 times using four different rates r0 / r1 / r2 / r3 respectively, determine the transmission rate, and update the basic rate at the same time;

[0011] Step 5, Normal Transmission Process: Send data packets using the transmission rate determined in Step 3;

[0012] Step 6: Calculate the frame transmission rate based on the ACK (acknowledgment message) feedback;

[0013] Step 7: At the end of the data slot, adjust the SNR low threshold value according to the frame transmission rate;

[0014] Step 8: Perform threshold correction on the SNR;

[0015] Step 9: When a new control time slot arrives, repeat steps 2 to 8 until the node has no more sending requests.

[0016] In step one, SNR is used. i Indicates selection of MCS i The low threshold value for transmission is set as SNR0 to th0, and the SNR gradient is δ, i.e., SNR1 = th0 + δ, SNR2 = th0 + 2 × δ, ..., SNR n =th0 + n × δ. Obtain an SNR-MCS mapping table.

[0017] In step three, the 802.11 RF rate configuration is achieved through index values. The MCS modulation and coding table is a representation proposed by 802.11 to characterize the communication rate of WLAN. MCS uses the factors affecting the communication rate as columns and the MCS indexes as rows to form a rate table. Therefore, each MCS index actually corresponds to a physical transmission rate under a set of parameters.

[0018] In step four, combining the basic rate r0 obtained in step three, four rate pairs are found: basic rate r0 / c0, high probe rate r1 / c1, low probe rate r2 / c2, and minimum available rate r3 / c3. The basic rate is obtained from the MCS-SNR mapping table; the high and low probe rates are the rates corresponding to the two MCS index values ​​adjacent to the basic rate; and the minimum available rate is the rate that guarantees successful transmission. During the sampling transmission process, at most (c0+c1+c2+c3) frames are sent, starting with the basic rate... If the transmission is successful after sending c0 times at the base rate r0, then the transmission rate is changed to the base rate r0. If the transmission is successful again, the base rate r1 is set as the transmission rate. If the transmission fails to be successful within c1 times using the base rate r0, then the transmission rate is changed to the base rate r0. If the transmission fails to be successful within c0 times using the base rate r0, then the transmission rate is changed to the base rate r2. If the transmission is successful, the base rate r2 is set as the transmission rate. If the transmission fails, then the lowest available rate r3 is set as the transmission rate.

[0019] In step six, the remaining frames are transmitted using the transmission rate determined in step four, and the frame delivery ratio (FDR) in a data slot is calculated. It is necessary to distinguish between collision-induced packet loss and channel-induced packet loss: For each frame transmitted, the total number of frames is t = t + 1; each time the RTS / CTS channel reservation fails, the reservation failure counter β = β + 1; each time a data frame is successfully transmitted, α = α + 1; then the FDR at this time is given by the following formula:

[0020]

[0021] In step seven, the online calibration mechanism used is as follows: For the FDR of the entire data slot statistics, the SNR of the control slot read is adjusted. When the SNR... <SNR i When FDR > 0.1, SNR i =SNR; when SNR <SNR i+1 When FDR > 0.9, SNR i+1 =SNR.

[0022] After updating the threshold, there may be a situation where the updated threshold value is greater than the threshold value corresponding to the next rate. Therefore, it is necessary to correct the monotonicity of the threshold in step eight. The correction process needs to ensure: 1) SNR i ≤SNR i+1 ;2)SNR i+1 ≤SNR i +δ. When correction is deemed necessary, δ is used as a reference value for vertical movement, i.e., SNR. afterCalibration =SNRbeforeCalibration ±δ.

[0023] Beneficial effects: Compared with the prior art, the present invention has the following advantages: 1) The method of the present invention can directly obtain the SNR information of the receiving end using the data time slot, avoiding the prediction bias introduced by the traditional algorithm based on predicting the SNR at the receiving node; 2) It can dynamically adjust the SNR threshold according to the frame transmission rate, which is highly dynamic and overcomes the problem that statistical algorithms cannot cope with rapidly changing channel conditions; 3) It can be applied to new mobile ad hoc network devices based on the separation of control plane and data plane, and can make reasonable use of control plane information to make decisions on the transmission rate of data time slots, thereby improving the system throughput. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating the stages of the adaptive modulation and coding selection method for mobile ad hoc networks based on the separation of control plane and data plane, as described in the embodiments.

[0025] Figure 2 This is a diagram of the control time slot situation frame structure involved in the embodiment;

[0026] Figure 3 This is a flowchart illustrating the transmission rate selection process during the sampling and transmission process described in the embodiment.

[0027] Figure 4 This is a diagram illustrating the calibration process of the online calibration mechanism involved in the embodiment. Detailed Implementation

[0028] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0029] The modulation and coding adaptive selection method for mobile ad hoc networks described in this embodiment mainly consists of nine stages: initializing the mapping table, obtaining the SNR, selecting the MCS, sampling and transmission process, normal transmission process, calculating the FDR, updating the SNR threshold, threshold correction, and time slot refresh. Figure 1 This is a flowchart of the stages of an adaptive modulation and coding selection method for mobile ad hoc networks based on the separation of the control plane and the data plane.

[0030] Step 1: Initialize the mapping table

[0031] Using a low threshold value indicates selecting the lowest SNR value for the current MCS. Each MCS corresponds to a low SNR threshold, i.e., when the SNR... i ≤SNR <SNR i+1 When selecting MCSi Set SNR0 to th0, and the SNR threshold corresponding to MCS0 to th0. A typical value for this threshold can be given based on actual measurements; a common typical value is 5 dB. Set the SNR gradient to δ, i.e., SNR1 = th0 + δ, SNR2 = th0 + 2 × δ, ..., SNR... n =th0 + n × δ, where the typical value of δ is 4 dB. Obtain an SNR-MCS mapping table that covers all MCSs.

[0032] Step 2: Obtain SNR

[0033] In the mobile ad hoc network constructed by this device, the relative position of each node is constantly changing. Therefore, the signal power received by each node from other nodes and the noise power of each node at each location are also changing in real time. In practical applications, each node needs to use a certain time interval. The data link layer of this device implements the separation of the control plane and data plane in a time-division manner, forming two relatively independent ad hoc virtual subnets. In the control time slot, the node receives situational frame information from neighboring nodes, merges it with the existing neighboring node information, and simultaneously senses the link status between the node and neighboring nodes, calculates the SNR, and fills it into the situational frame that the node will send. In the control plane subnet, a priority CMSA mechanism is used to ensure that each node has the opportunity to send situational information. In the data plane subnet, the same CSMA mechanism is used, and the situational information from the control plane can be used to form a certain "network order" and reduce the throughput decrease caused by packet errors and collisions.

[0034] In mobile ad hoc networks, a receiver often has multiple neighboring nodes, but the SNR values ​​of these neighboring nodes are not necessarily the same. Therefore, a specific situational frame structure is needed to distinguish the SNR between different links, as shown in the following format: Figure 2 As shown.

[0035] When a neighboring node receives the frame, it extracts the MAC information from the frame and matches it with the neighboring node information stored locally. If a match is found, the corresponding SNR information is updated; otherwise, a new neighboring node information is created and stored locally. At the start of a data slot, when a data frame is sent to the destination MAC address, the node first iterates through the locally recorded neighboring node MAC and SNR mapping information, finds the destination MAC address entry, and reads the corresponding SNR information.

[0036] Step 3: Select MCS

[0037] The configuration of 802.11 radio frequency rates is achieved through index values. The Modulation-Coding Table (MCS) is a representation proposed by 802.11 to characterize the communication rate of a WLAN. The MCS uses the factors affecting the communication rate as columns and the MCS indexes as rows, forming a rate table. Therefore, each MCS index actually corresponds to a physical transmission rate under a set of parameters.

[0038] Based on the neighbor node information stored locally, obtain the destination MAC address of the data frame, the corresponding SNR information of the receiving end, traverse each item of the SNR-MCS mapping table, and find the MCS corresponding to the SNR as the basic rate.

[0039] Step 4: Sampling and Transmission Process

[0040] The specific sampling and transmission process is shown in the attached figure. Figure 2 As shown. Combining the basic rate r0 obtained in the previous step, four rate pairs can be derived, expressed in the form of rate / retry count. They are: basic rate r0 / c0, high probe rate r1 / c1, low probe rate r2 / c2, and minimum available rate r3 / c3. The basic rate is obtained from the SNR read from the locally stored neighbor information in the mapping table established in the previous step. The high and low probe rates are the rates corresponding to the two MCS index values ​​adjacent to the MCS index value corresponding to the basic rate. The minimum available rate is the rate that guarantees successful transmission, which is generally selected as MCS1. During the sampling transmission process, at most (c0+c1+c2+c3) frames are transmitted. First, c0 transmissions are performed at the basic rate r0. If successful, a higher probe rate r1 is used to transmit c1 times. If successful again, the higher probe rate r1 is selected as both the transmission rate and the basic rate. If transmission fails within c1 times using the higher probe rate r1, the basic rate r0 is used as both the transmission rate and the basic rate. If transmission fails within c0 times using the basic rate r0, a lower probe rate r2 is used to transmit c2 times. If successful, the lower probe rate r2 is set as both the transmission rate and the basic rate. If transmission fails, the lowest available rate r3 is used as the transmission rate, and the basic rate is set to max(r2-1, r3), representing the larger of the rate corresponding to the MCS index value minus one (r2) and the rate corresponding to r3. During the sampling transmission process, the typical values ​​for c0 / c1 / c2 / c3 are (2, 4, 4, 2).

[0041] Step 5: Sampling and Transmission Process

[0042] This process uses the transmission rate r determined in the previous step to transmit the data frames that need to be sent in the data time slot.

[0043] Step 6: Calculate FDR

[0044] While transmitting frames in a data time slot, the Frame Rate of Transmission (FDR) of frames transmitted in that data time slot is calculated. This calculation process needs to distinguish between collision packet loss and channel packet loss: For each frame transmitted, the total number of frames is t = t + 1; each time an RTS / CTS channel reservation fails (i.e., an RTS reservation is sent but a CTS response is not received), the reservation failure counter β = β + 1; each time a data frame is successfully transmitted (i.e., an ACK is received from the receiver), α = α + 1; then the FDR at this time is given by the following formula:

[0045]

[0046] Step 7: Update the SNR threshold

[0047] The specific update mechanism is as follows: Figure 3 As shown. The threshold is updated by combining the FDR obtained from data slot statistics with the SNR of the control slot read. When the SNR... <SNR i When FDR > 0.1, SNR i =SNR, that is, updating the SNR threshold corresponding to the current MCS to the SNR read from the control slot; when SNR <SNR i+1 When FDR > 0.9, SNR i+1 =SNR, which updates the lower threshold of SNR of the current MCS level to the SNR read from the control slot; when SNR and FDR do not meet any of the above conditions, the threshold value remains unchanged.

[0048] If MCS=5 is currently used for transmission, then SNR5=22 and SNR6=26. If the SNR read from the control slot is 24 and FDR=0.92>0.9, then SNR6 is updated to 24. If, after decision, MCS=6 is used for transmission and FDR=0.35>0.1, then SNR6 is updated to 24. If, at this time, MCS=5 is used for transmission, but FDR=0.75<0.9, the threshold value will remain unchanged.

[0049] Step 8: Threshold Correction

[0050] After updating the threshold, there may be cases where the updated threshold value is greater than the threshold value corresponding to the next higher MCS. Therefore, it is necessary to correct the monotonicity of the threshold. The correction process needs to ensure: 1) SNR i ≤SNR i+1 ;2)SNR i+1 ≤SNR i +δ. When correction is deemed necessary, δ is used as a reference value for vertical movement, i.e., SNR. afterCalibration =SNR beforeCalibration ±δ. Where δ is the gradient of SNR.

[0051] Step 9: Time Slot Refresh

[0052] After the above steps, the low threshold value of SNR corresponding to MCS is updated. When a new control time slot arrives, this node rereads the SNR value of the receiving node from the situation frame. According to the SNR-MCS mapping relationship, the rate corresponding to MCS is reselected as the basic rate. The transmission rate is selected using the sampling transmission process and transmitted during the normal transmission process. The threshold is updated by combining SNR and FDR. Steps two to eight are repeated to realize the modulation and coding adaptive process.

[0053] The embodiments described above are merely preferred embodiments of the present invention, and while their descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for modulation and coding selection in mobile ad hoc networks based on the separation of control and data planes, characterized in that, For mobile ad hoc network devices that employ a separation of control plane and data plane, the following steps are included: Step 1: Construct a mapping table within the mobile ad hoc network device for the current link's low SNR threshold and the MCS index values ​​of all supported modulation and coding schemes; Step 2: Read the latest updated SNR from the transmitting node to the receiving node from the situation frame received in the control time slot; Step 3: Traverse the mapping table established in Step 1 and select the MCS index value corresponding to the current SNR as the base rate. Step 4, Sampling and Transmission Process: Transmit c0 / c1 / c2 / c3 times using four different rates r0 / r1 / r2 / r3 respectively, determine the transmission rate, and update the basic rate at the same time. Step 5, Normal Transmission Process: Send data packets using the transmission rate determined in Step 3; Step 6: Calculate the frame transmission rate based on the feedback of ACK (acknowledgment) messages; Step 7: At the end of the data slot, adjust the SNR low threshold value according to the frame transmission rate; Step 8: Perform threshold correction on the SNR; Step 9: When a new control time slot arrives, repeat steps 2 to 8 until the node has no more sending requests.

2. The method of claim 1, wherein the method is characterized by, In step one, use SNR i Indicates selection of MCS i The low threshold value for transmission is set as SNR0 to th0, and the SNR gradient is δ, i.e., SNR1 = th0 + δ, SNR2 = th0 + 2 × δ, ..., SNR n =th0+n×δ.

3. The method of claim 1, wherein the method is characterized by, In step three, the configuration of the 802.11 radio frequency rate is achieved through index values; the MCS modulation and coding table is a representation proposed by 802.11 to characterize the communication rate of WLAN; the MCS uses the factors of interest affecting the communication rate as columns of the table and the MCS index as rows to form a rate table; therefore, each MCS index actually corresponds to the physical transmission rate under a set of parameters.

4. The method of claim 1, wherein the method further comprises: In step four, combining the basic rate r0 obtained in step three, four rate pairs are found: basic rate pair r0 / c0, high-probing rate pair r1 / c1, low-probing rate pair r2 / c2, and minimum available rate pair r3 / c3. The basic rate is obtained from the MCS-SNR mapping table. The high and low probing rates are the rates corresponding to the two MCS index values ​​adjacent to the basic rate. The minimum available rate is the rate that guarantees successful transmission. During the sampling transmission process, at most (c0+c1+c2+c3) frames are transmitted. First, c0 frames are transmitted at the basic rate r0. If the transmission is successful, c1 frames are considered to be transmitted using the high-probing rate r1. If the transmission is successful again, the high-probing rate r1 is selected as the transmission rate. If the transmission fails within c1 frames using the high-probing rate r1, the basic rate r0 is used as the transmission rate. If the transmission fails within c0 frames using the basic rate r0, the low-probing rate r2 is considered. Transmit c2 times. If the transmission is successful, set the low probe rate r2 to the transmission rate. If the transmission fails, the lowest available rate r3 will be used as the transmission rate.

5. The method of claim 1, wherein the method is characterized by, In step six, the remaining frames are sent using the transmission rate determined in step four, and the frame rate of return (FDR) in a data time slot is calculated. It is necessary to distinguish between collision packet loss and channel packet loss: the total number of frames is t = t+1 for each frame sent; the reservation failure counter β = β+1 each time the RTS / CTS channel reservation fails; and α = α+1 each time a data frame is successfully sent. Then the FDR at this time is given by the following formula: FDR = α / (t-β) × 100%.

6. The method of claim 1, wherein the method is characterized by, In step seven, the online calibration mechanism used is as follows: for the FDR of the entire data slot statistics, the SNR of the control slot read is adjusted when... SNR < SNR i  and FDR > 0.1, SNR i = SNR; when SNR < SNR i+1 and FDR > 0.9, SNR i+1 = SNR.

7. The method of claim 1, wherein the method is characterized by, In step eight, monotony of the threshold is corrected, and the correction process needs to ensure that: 1) SNR i ≤ SNR i+1 ; 2) SNR i+1 ≤ SNR i + δ; When it is determined that correction is needed, δ is used as a reference value for vertical movement, i.e., SNR. afterCalibration =SNR beforeCalibration ±δ.

Images