CN104602280B - The Distributed admission control method of low-power consumption body area network based on adaptive polling - Google Patents

Abstract

The invention provides a distributed access control method of a low-power body area network based on adaptive polling, and belongs to the field of wireless body area networks. According to the change of channel transmission conditions, the present invention adaptively adjusts the polling period T to achieve energy optimization, and by changing the polling period and data transmission rate, realizes the optimal energy consumption of nodes per unit time, thus reducing the power consumption of the network. consumption and prolong the life of nodes. The present invention improves the deficiencies of existing algorithms in practical applications, considers the activity status of nodes in the polling cycle as a whole, and measures the energy consumption of nodes comprehensively. Starting from the actual application scene, it proves that self-adaptive adjustment of system parameters can obtain Algorithm performance optimization.

Description

Distributed access control method for low-power body area network based on adaptive polling

technical field

The invention is proposed aiming at the wireless low-power body area network technology, and in particular relates to an access control method of the low-power body area network based on adaptive polling.

Background technique

Wireless Body-Coupled Communication (WBCC) is a new type of communication method centered on the human torso, which realizes data exchange between the inside and outside of the human body, the human body and surrounding fixed or mobile terminals within a certain range. Wireless Sensor Networks (WSNs). Compared with traditional wireless communication methods, this technology has low power consumption, high confidentiality, and less damage to the human body, and it can well solve the problem of traditional wireless communication that reduces communication efficiency during multi-point access. It is expected to achieve comparable results. Better application effect of traditional wireless communication methods.

Wireless human body area network technology has very broad application prospects in medical treatment, navigation and positioning, personal multimedia and military, such as intelligent diagnosis, treatment, intelligent transportation, intelligent clothing, etc., and the realization of these applications is based on the low-power wireless sensor network. Based on the development, it is of great significance to design a low-power access control method.

The existing wireless sensor network access control methods are mainly divided into two categories: synchronous method and asynchronous method. The synchronous method mainly includes S-MAC, T-MAC, etc. The most distinctive feature of this type of method is that all nodes sleep periodically. And awake, while the asynchronous method uses a long signal frame longer than the sleep time to wake up the receiving node to realize signal transmission. Representative algorithms include T-MAC, X-MAC, Wise-MAC, etc. However, in the ISM frequency band, these methods are all based on the radio frequency physical layer, and the problems of human shadow and low energy efficiency caused by this have not been well solved.

Later, foreign scholars tried to use the human body as a communication channel, and proposed the concept of Body-Coupled Communication (BCC), which improved communication reliability and energy efficiency. Among them, Steven Corroy et al. proposed a new asynchronous access control method using body-coupled channels as the physical layer: Ada-MAC algorithm. After the simulation experiment comparison of NS-2 network simulator, it is found that under the same human body coupling communication network parameters, this method has better reliability, lower delay and higher energy efficiency than other MAC algorithms.

However, the above method also has several deficiencies: first, it only considers the energy of the node sending the pilot stage, but does not consider the energy consumed by the node in all activities; second, it does not take into account the different rates of different data streams, because in In practical applications, different data types are transmitted on the human body channel, and the data transmission rate is also different.

Contents of the invention

In view of the shortcomings of the above method, the present invention proposes an access control method for low power consumption body area network based on adaptive polling. The technical scheme of the present invention is as follows.

A distributed access control method for low-power body area networks based on adaptive polling, according to changes in channel transmission conditions, adaptively adjust the polling period T to achieve energy optimization, and the channel transmission conditions include transmission data types and channel transmission rate, specifically including the following steps:

Step 1: Establish the access activity model per unit time: In the body-coupled human body channel, the active states of nodes include four types: polling, dormancy, sending and receiving; the polling state means that the node periodically monitors the channel state, as The node prepares for sending data or receiving data; the dormant state refers to a state in which the node neither sends data nor receives data. In this state, the node will not receive information or data transmitted in the channel, nor can it send data to Other nodes; the sending state is the process in which the node uses the channel to send data or information to other nodes; the receiving state refers to the process in which the node receives the information or data sent by other nodes; in the body-coupled human body channel, all nodes will periodically take turns Polling channel, polling is the process of self-adaptive wake-up of the node, and it is the beginning of the transition of the node in all states;

Step 2: Establish a node energy consumption model per unit time: the total energy consumption (E) of a node within a unit time is defined as the energy consumed by the polling state during this period (E _poll ), and the energy for sending the state (E poll ) _tx ), the sum of the energy in the receiving state (E _rx ) and the energy in the dormant state (E _sleep );

Step 3: Find the optimal energy consumption of the node, that is, the minimum energy consumption of the node; introduce the parameter R _data to represent the data transmission rate, and the node determines R _data according to the type of data flow transmitted in the network, in order to determine the minimum energy consumption per unit time The optimal polling period T in the case of That is, the optimal polling period T of the node at a specific data transmission rate is obtained.

In the above step 1, there are three possible states of nodes in a polling cycle as follows:

Scenario 1:

1) poll the channel and check that the channel is idle;

2) The node does not need to send or receive data, and the node goes into a dormant state at this point, waiting for the next polling;

Scenario 2:

1) poll the channel and check that the channel is idle;

2) The channel is idle, the source node needs to send data to the destination node, the source node starts sending pilot frames to the destination node, and the source node enters the sending state;

3) The nodes in the network periodically poll the channel, detect the pilot frame sent by the source node, and determine the destination node through the information in the pilot frame, and at this time, the destination node sends an RTR packet To the source node, notify the source node to start sending data packets;

4) After the source node receives the RTR packet sent by the destination node, it starts sending data until the end of sending data;

5) After the data transmission is completed, the destination node sends an ACK packet to the source node to confirm the end of the data transmission process, and after the source node receives the ACK packet sent by the destination node, the source node and the The destination nodes all enter the dormant state, waiting for the next polling;

Case three:

1) poll the channel and check that the channel is occupied;

2) The node receives the pilot frame in the channel, determines itself as the destination node of the pilot frame according to the purpose information of the pilot frame, and enters the receiving state;

3) The node sends an RTR packet to the source node that sends the pilot frame, and the source node starts sending data after receiving the RTR packet sent by the node;

4) The node starts to receive data until the end of data reception, and the node sends an ACK packet to the source node to confirm the end of data sending;

5) The source node receives the ACK data packet, and the node and the source node enter a dormant state, waiting for the next polling.

In the above control method, in a polling cycle, the polling state always occurs first, and the node always polls the channel first before sleeping, sending data or receiving data, and monitoring whether the channel is idle; the present invention establishes a node in the unit time Based on the energy model within, and based on this, the expression of the energy consumption of the node in unit time is calculated, which proves theoretically that the energy consumption of the node will change with the change of the polling cycle, and under the condition of the optimal polling cycle Reach the lowest node energy consumption.

In the above step three, according to the relationship between the energy consumption per unit time of the node and the polling period, the influencing parameters of the optimal polling period are obtained, so that in the body-coupled human body network, the polling period T is adaptively adjusted to achieve the energy Optimal; optimal polling cycle:

In the formula, P _poll >P _sleep means that the polling power is greater than the sleep power. It can be seen from formula (22) that when the channel is in normal operation, the polling cycle T and the single polling time L _poll-once , the data transmission rate R _data and the guide The frequency parameter p is related, and the basic relationship is as follows: Under the condition of a single condition change, the longer the duration L _poll-once of a single poll, the longer the polling cycle T; the larger the data transmission rate R _data , the polling cycle T is shorter; and because the pilot parameters p>1, k>1, Then as p is larger, the polling period T is longer.

Compared with the prior art, the present invention has the following advantages and technical effects: the present invention realizes the optimal energy consumption of nodes per unit time by changing the polling cycle and data transmission rate, thereby reducing the power consumption of the network and prolonging the life of nodes . Under the action of the above mechanism, the present invention can realize the optimization of node energy consumption as a whole. In practical applications, such as in the medical field, nodes adjust the transmission rate according to the type of data transmitted to obtain the optimal polling cycle T, reduce network power consumption, extend node life, and achieve performance optimization.

Description of drawings

Figure 1 is a schematic diagram of the activities of nodes in a unit time in the example.

Fig. 2 is a schematic diagram of the activities of the node in case 1 in the example.

Fig. 3 is a schematic diagram of the activities of the node in situation 2 in the example.

Fig. 4 is a schematic diagram of the activities of the node in case 3 in the example.

Detailed ways

The present invention will be further described in detail through specific embodiments below in conjunction with the accompanying drawings, but the implementation of the present invention is not limited thereto.

Firstly, the activity state model of the nodes in the human body coupling communication network channel is analyzed.

In the body-coupled human body channel, there are four main active states of nodes: polling, sleeping, sending and receiving. The polling state means that the node periodically monitors the channel state to prepare for the node to send data or receive data. The dormant state refers to a state in which the node neither sends data nor receives data. In this state, the node will not receive information or data transmitted in the channel, nor can it send data to other nodes. The sending state is the process in which a node uses a channel to send data or information to other nodes. The receiving state refers to the process in which a node receives information or data sent by other nodes. In body-coupled human body channel, all nodes poll the channel periodically, so polling is the process of node adaptive wake-up, which is the beginning of node transition in all states.

As shown in Figure 1, it is the flow chart of the node’s activity status per unit time. Without considering retransmissions, conflicts, etc., it is assumed that the node polls the channel at a period of T, and sends and receives data according to whether the channel is occupied or not. The need to transition in various states of sleep state, sending state and receiving state.

In a polling cycle, the status of the node may change in the following three situations:

1. Situation 1 (as shown in Figure 2): The node polls and detects the channel, the channel is idle, and the node does not need to send or receive data, and directly enters the sleep state, waiting for the next poll;

2. Situation 2 (as shown in Figure 3): Node 1 (source node) polls the channel, the channel is idle and needs to send data to node 2 (destination node), then the node 1 occupies the channel and enters the sending state, and sends a pilot frame to node two. At this time, if the node 2 is in the sleep state, the node 1 will continue to send pilot frames until the node 2 enters polling after the sleep state and receives the pilot frame transmitted by the node 1. After the node two receives the pilot frame transmitted by the node one, after confirming that the node is the target node of the pilot frame, it immediately returns an RTR (Ready To Receive) packet to the node one. As soon as the node confirms, it starts sending data. When the node 1 receives the RTR packet from the node 2, the sending connection between the two nodes is established, the node 1 starts sending data packets, and the node 2 starts receiving data packets until the sending process ends. Finally, the node 2 sends back an ACK confirmation packet until the node 1 confirms that the sending process is over, and after the node 1 receives the ACK confirmation frame, the sending connection between the two nodes is disconnected, the channel is unoccupied, and the node 1 Both node 1 and node 2 enter a dormant state and wait for the next polling.

3. Situation three (as Fig. 4): described node one wakes up and starts polling channel, finds that channel is occupied, and detects the pilot frame from described node two, and described node one according to described pilot frame The target address information confirms that the current node is the target node of the pilot frame. Then send the RTR packet to the node two to confirm and start sending data. After the node two receives the RTR packet sent back by the node one, the sending connection between the two nodes is set up, and the node two starts sending data The node 1 starts to receive the data packet until the data transmission ends, and then the node 1 sends an ACK data packet to the node 2 to confirm the end of sending the data packet, and the node 2 receives the ACK data packet sent by the node 1. After the ACK data packet, the data connection between the two nodes is disconnected, and the channel is unoccupied, and the node one and the node two start to enter the dormant state, waiting for the next polling; if the node one receives the After detecting the pilot frame, it finds that it is not the destination node of the pilot frame, and then directly enters the dormant state, waiting for the next polling.

Next, the energy consumed by a node per unit time will be explained.

Define a node energy consumption model per unit time: define the total energy consumption per unit time as the sum of the energy consumed in each state per unit time, that is, the total energy consumption E per unit time is equal to the energy in the polling state per unit time E _poll , the sum of the energy E _tx of the sending state, the energy E _rx of the receiving state, and the energy E _sleep of the sleeping state, namely:

E＝E _poll +E _tx +E _rx +E _sleep (1)

According to the definition of energy, it can be obtained that the energy consumed by a node in each state per unit time is equal to the product of the power (P) in each state and the state duration (L). If P _poll , P _tx , P _rx , P _sleep represents the power of the node in the polling state, sending state, receiving state and sleeping state respectively, let L _poll , L _tx , L _rx , L _sleep respectively represent the duration of the node in the polling state, sending state, receiving state and sleeping state time, there are:

E _poll ＝P _poll ·L _poll (2)

E _tx = P _tx L _tx (3)

E _rx ＝P _rx ·L _rx (4)

E _sleep = P _sleep L _sleep (5)

In unit time, the default conditions are:

L _poll +L _tx +L _rx +L _sleep = 1 (6)

(1)E _poll

Let T denote the polling period, then Indicates the number of polls per unit time. If L _poll-once is used to represent the time of a single poll, then there is

Available:

(2) E _tx , E _rx

Since in the whole access process, sending and receiving data are the two most important processes, and the sending and receiving processes are intersected with each other, for example, when a node sends data, it is also accompanied by receiving RTR packets and ACK packets. Process; Correspondingly, when receiving data, it is also accompanied by the process of sending RTR packets and ACK packets. Therefore, the process of sending and receiving data is combined to analyze its energy consumption.

In the body-coupled human body channel, the received RX power and the transmitted TX power are asymmetrical. Since the transmit power of the transceiver in body-coupled communication is limited and the incoming power of the receiver is very low, in order to achieve full-body coverage, the receive power P _rx is higher than the transmit power P _tx , that is, (in the existing solution, k The value often exceeds 4, even reaches 10):

P _rx = kP _tx (9)

Introduce the variable R _data , which is defined as the average number of data packets sent by a node per unit time. When the data packet length is determined, the greater the data transmission rate, the greater the R _data correspondingly, so R _data can be used to represent the data transmission rate .

Let L _μ represent the time required to transmit a pilot frame, L _L represents the interval time between adjacent pilot frames, L _μ should be greater than L _L , then assume:

L _μ = pL _L (10)

The time L _p to send a pilot is equal to the time to send n pilots and n-1 pilot intervals, then:

L _p ＝nL _μ +(n-1)L _L (11)

In the second situation (as shown in Figure 3), the duration L _tx of the node transmission state includes the time for sending pilot frames (R _data ·nL _μ ), the time for sending data packets (R _data ·L _data , L _data represents the time for transmitting data) and the time for sending two confirmation frames (2m·R _data ·L _L , m represents the average number of nodes sending data to the node per unit time) under the third situation (as shown in Figure 4), then Have:

L _tx =R _data ·nL _μ +R _data ·L _data +2m·R _data ·L _L (12).

The duration L _rx of the node being in the receiving state in the situation three (as shown in Figure 4) includes the time (R _data ·(n-1 ) L _L ), the time of receiving two confirmation frames (2m·R _data ·L _L ) and the pilot frame (m·R _data ·1.5L _μ ) received in the third case (as shown in Figure 4), m represents the unit The average number of nodes that send data to the node within the time, 1.5L _μ means that a complete pilot frame can be obtained only after receiving 1.5 pilot frames on average) and the time of the data packet (m · R _data · L _data ), then there is :

L _rx ＝R _data ·(n-1)L _L +2m ·R _data ·L _L +m ·R _data ·1.5L _μ +m ·R _data ·L _data (13)

The energy consumed in the process of sending a pilot frame is equal to the sum of the energy consumed by the sending pilot in the sending state and the energy in the receiving state between the receiving pilot, that is:

E _p =P _tx ·R _data ·nL _μ +P _rx ·R _data ·(n-1)L _L (14)

Substituting into the formula (9), then:

E _p ＝P _tx ·R _data [nL _μ +k·(n-1)L _L ] (15)

According to the conclusions drawn in the document "Low Power Medium Access Control for Body-Coupled Communication Networks", the average value of energy consumed by sending pilot frames is:

make Then there are:

In a poll, there should be

L _S ＝TL _poll-once (18)

Then you can get:

(3) E _sleep

From the formula (6) can get:

L _sleep ＝1-L _poll -L _tx -L _rx (20)

E _sleep ＝P _sleep ·(1-L _poll -L _tx -L _rx ) (21)

To sum up, the total energy consumed per unit time is:

in,

In order to achieve optimal energy, it is necessary to determine the optimal value of the polling cycle, assuming m, k, p, P _poll , P _tx , P _rx , P _sleep and L _poll-once , L _L , L _μ , R _data , etc. Treated as a constant, find Then, the optimal polling cycle is obtained:

From the formula (22), it can be obtained that a prerequisite for the optimal polling period T is that P _poll >P _sleep , that is, the polling power is greater than the sleep power, which is in line with the general network law, because in the sleep state, the node does not Visible activity, therefore, sleep power is less than polling power. It can be known from formula (22) that when the channel is running normally, the polling period T is related to the single polling duration L _poll-once , the data transmission rate R _data and the pilot parameter p. The basic relationship is as follows: under the condition of only considering the change of a single condition, the longer the duration L _poll-once of a single poll, the longer the polling period T; the larger the data transmission rate R _data , the shorter the polling period T; And because the pilot parameters p>1, k>1, Then as p is larger, the polling period T is longer. It can be seen that according to changes in channel transmission conditions (transmission data type, channel transmission rate, etc.), adaptively adjusting the polling cycle T can achieve energy optimization, thereby achieving the purpose of reducing network power consumption and extending node life.

Claims (3)

1. The distributed access control method of the low-power-consumption body area network based on the self-adaptive polling is characterized in that a polling period T is self-adaptively adjusted to achieve the optimization of energy according to the change of channel transmission conditions, wherein the channel transmission conditions comprise a transmission data type and a channel transmission rate, and the method specifically comprises the following steps:

the method comprises the following steps: establishing an access activity model in unit time: in the body-coupled human body channel, the activity states of the nodes comprise polling, dormancy, sending and receiving; the polling state refers to that the node periodically monitors the channel state and prepares for sending data or receiving data by the node; the dormant state refers to a state in which the node neither sends data nor receives data, and in this state, the node does not receive information or data transmitted in a channel, and cannot send data to other nodes; the sending state is a process that the node sends data or information to other nodes by using a channel; the receiving state refers to a process that the node receives information or data sent by other nodes; in the body-coupled human body channel, all nodes can poll the channel periodically, wherein the polling is the process of self-adaptive awakening of the nodes and is the beginning of the conversion of the nodes in all states;

step two: establishing a node energy consumption model in unit time: the total energy consumption (E) of a node in a unit of time is defined as the energy (E) consumed by the polling state during the period of time_poll) Energy of the transmitting state (E)_tx) Receiving energy of state (E)_rx) And energy (E) of the sleep state_sleep) The sum of (a);

step three: solving the optimal energy consumption of the node, namely the minimum value of the energy consumption of the node; introduction of parameter R_dataRepresenting the data transmission rate, the node determining R according to the type of data stream transmitted in the network_dataIn order to determine the optimal polling period T in the case of the minimum energy consumption per unit time, a derivation algorithm is usedObtaining the optimal polling period T of the node under a specific data transmission rate; obtaining an influence parameter of the optimal polling period according to the relation between the energy consumption of the node in unit time and the polling period, thereby adaptively adjusting the polling period T to optimize the energy in the body-coupled human body network; optimal polling period:

in the formula, P_poll＞P_sleepI.e. the polling power is greater than the sleep power, P_txExpressing the power of the node in the transmitting state, the formula (22) shows that the channel runs normally and turnsPolling period T and single polling duration L_poll-onceData transmission rate R_dataAnd the pilot parameter p has a relationship, the basic relationship of which is as follows: duration L of single polling under single condition change_poll-onceThe longer the polling period T is; data transmission rate R_dataThe larger the polling period T is; and because of the pilot parameter p>1,k>1,The polling period T is longer as p is larger.

2. The distributed access control method for a low power consumption body area network based on adaptive polling according to claim 1, wherein in step one, the possible states of the nodes existing in one polling period are as follows:

the first situation is as follows:

1) polling the channel, and checking whether the channel is idle;

2) the node does not need to send or receive data, and the node is switched into a dormant state to wait for next polling;

case two:

1) polling the channel, and checking whether the channel is idle;

2) when a channel is idle, a source node needs to send data to a destination node, the source node starts to send a pilot frequency frame to the destination node, and the source node enters a sending state;

3) a node in a network periodically polls a channel, detects a pilot frequency frame sent by a source node, determines a target node according to information in the pilot frequency frame, and at the moment, the target node sends an RTR packet to the source node to inform the source node of starting to send a data packet;

4) after receiving the RTR packet sent by the destination node, the source node starts to send data until the data sending is finished;

5) after data transmission is finished, the destination node transmits an ACK packet to the source node, the data transmission process is confirmed to be finished, and after the source node receives the ACK packet transmitted by the destination node, the source node and the destination node both enter a dormant state and wait for next polling;

case three:

1) polling channels, and checking channel occupation;

2) the node receives a pilot frequency frame in a channel, determines a destination node which is a pilot frequency frame according to destination information of the pilot frequency frame, and enters a receiving state;

3) the node sends an RTR packet to the source node which sends a pilot frequency frame, and the source node starts to send data after receiving the RTR packet sent by the node;

4) the node starts to receive data until the data reception is finished, and the node sends an ACK data packet to the source node to confirm the data transmission is finished;

5) and the source node receives the ACK data packet, and the node and the source node enter a dormant state to wait for next polling.

3. The distributed access control method for low power body area network based on adaptive polling according to claim 1, wherein in a polling period, the polling status always occurs first, and the node always polls the channel before sleeping, sending data or receiving data to monitor whether the channel is idle.