CN102711211B - 6LoWPAN-based energy-saving data transmitting method for wireless sensor network - Google Patents

Abstract

The invention provides a 6LoWPAN (6Low-power Wireless Personal Area Network)-based energy-saving data transmission method for a wireless sensor network. An optimal TAN (Temporary Assembling Node) needs to be found between an information source and an information sink; firstly, the information source divides an IPv6 (Internet Protocol Version 6) into fragmentations, then transmits the fragmentations to the TAN one by one through a method of MUR, namely Mesh-under Routing, of the 6LoWPAN protocol; and the TAN assembles together the received fragmentations to the original IPv6 data packet, then fragments the IPv6 data packet, and transmits the fragmentations to the information sink node one by one through the MUR. Therefore the total energy consumption of the node transmitting data packet is reduced. The method of finding the TAN is the key of the invention, and the core of the invention is formula (19). By adopting the method provided by the invention, the energy consumption of the node transmitting data packet, the retransmission times of the data packet and the communication delay-time in the wireless sensor network are all reduced.

Description

A kind of wireless transducer network energy saving data transferring method based on 6LoWPAN

Technical field

The present invention relates to a kind of Energy-saving Data transmission method, be applicable to the wireless sensor network based on 6LoWPAN.

Technical background

Wireless sensor network is conventionally using IEEE 802.15.4 as communication standard, and it is widely used in the fields such as environmental observation, Smart Home, medical treatment and nursing, traffic monitoring.In IEEE 802.15.4 standard, the network equipment has the features such as low-power consumption, low data transmission rate, short communication distance, short packages length.For example, this standard specifies that link layer MTU MTU is that the size of Maximum Transmission Unit is 127 bytes; The wireless device maximum data transfer rate that works in 2.4GHz frequency range is 250kbps; Except the unique extended address in the whole world of 64, also support the short address form of 16.

IPv6 is that IETF is that the IP agreement for substituting current edition 4 that Internet Engineering task groups designs is the IP next generation protocol of IPv4.The address size of IPv6 is 128, and the minimum MTU of IPv6 packet reaches 1280 bytes.Because the MTU of IEEE 802.15.4 data link layer can only reach 127 bytes, so when IPv6 packet transmits in the wireless sensor network based on IEEE 802.15.4 standard, packet burst need to be transmitted.On August 1st, 2007, IETF drafts and has issued RFC 4919, has proposed 6LoWPAN agreement and has transmitted IPv6 packet to be supported in the wireless personal area network based on IEEE 802.15.4 standard.The main feature of 6LoWPAN has three: packet burst, header suppression and Level 2 Forwarding packet.It is adaptation layer that 6LoWPAN adds an adaptation layer between network layer and data link layer.Like this, route both can be carried out in network layer, also can carry out at newly-increased adaptation layer.The former is called ROR is route-over routing; It is mesh-under routing that the latter is called MUR.

Before the present invention, relevant invention has:

On October 18th, 2010, the application of Changshu Institute of Technology Wang Xiao nanmu patent " the nested mobile network routing optimization method of a kind of 6LoWPAN ".In this invention, mobile network node is set up tunnel without the mobile router agency through place mobile network and can be communicated by letter with communication node, mobile network node can according to the IPv6 address of object communication node accurately judge this communication node whether with mobile network node in same 6LoWPAN network, thereby set up the optimum routed path that arrives object communication node, shorten data transfer delay.

On November 23rd, 2010, the application of Changshu Institute of Technology Wang Xiao nanmu patent " a kind of 6LoWPAN wireless sensor network routing method ".The method is divided into global function sensor node and partial function sensor node two classes the node of wireless sensor network, in addition, also the IPv6 address of sensor node is divided into overall route prefix and sensor node ID two parts.Global function node comprises a routing table, thereby data communication is realized in the path that sensor node is set up arrival destination node by routing table.In the time that the global function sensor node in path lost efficacy, can automatically realize route repair function by routing table.

On January 26th, 2011, the auspicious patent " a kind of tree-shaped method for routing of finding based on 6LoWPAN neighbours " of having applied for of the Hao Jun of Wuhan Research Institute of Posts & Telecommunications.In the time that 6LoWPAN sensor node adds network, the method is set up the root node of tree topology, and essential information to their configuration networks.If occur, leaf node lost efficacy, left, when one of three kinds of situations of movement, trigger route repair process.The method, without the extra routing protocol packet of sending and receiving, has reduced energy consumption and the routing cost of whole network.

On August 10th, 2011, the yellow little red people such as grade of Beijing University of Post & Telecommunication applies for a patent " 6LoWPAN network is towards the TCP header compression method of http protocol ".The method first completes the compression of IP stem and IP extension header at adaptation layer, then according to the mode of header-compressed structure, coding and queue, TCP stem is compressed, and has ensured the compatibility of compression method and the simplicity that operation realizes.The 6LoWPAN datagram forming is sent to data link layer again, through data link layer and physical layer, frame is sent to receiving terminal.The method has reduced the data fragmentation of data link layer, has improved communication efficiency.

As previously mentioned, 6LoWPAN network has added adaptation layer between the IP of original ICP/IP protocol stack layer and MAC layer.What routing decision was completed by IP layer is called ROR, and routing decision is at the MUR that is called that adapts to complete layer by layer.

As Fig. 1, when a paths is made up of multiple nodes, and link between node is during all based on IEEE 802.15.4 standard, in ROR and these two kinds of routing algorithms of MUR, packet from source node A ₀send to destination node A _nprocess be different.In ROR, packet is first by source node A ₀be divided into multiple bursts, each burst transmits with IP agreement, with node A _niP address as the destination address of packet head.Source node A ₀select node A according to routing table ₁for down hop, all bursts are sent to next-hop node A ₁.Node A ₁receiving after all bursts, these bursts are assemblied into original packet, give network layer, after the object IP address of network layer sense data packet header, make routing decision according to routing table and select A ₂for down hop, then, the packet after assembled again burst, and all bursts are passed to next-hop node A ₂, node A ₂receiving after all bursts, these bursts are assemblied into original packet.The rest may be inferred, until packet arrives destination node A _n.

The data transfer of MUR is different from ROR.In MUR, route is carried out at adaptation layer: first packet is fragmented, and all bursts are that Local-link address forwarded hop-by-hop is to destination node A according to local link address _n.In IPv6, the prefix of local link address is FE80: :/64.Destination node A _nafter receiving all bursts, these bursts are assemblied into source node A ₀the packet sending.

No matter be ROR routing policy, or MUR routing policy, if lost the one or more bursts in packet on transmission link, all bursts of this packet all need to retransmit so.But the former only retransmits on the link of a jumping, and the latter need to retransmit end to end.Therefore, in the time that information source is excessive to the number of links number on compare Chang Ji path in path between the stay of two nights, the data packet retransmission number of times of MUR will become very large, thus the energy of waste node.

Summary of the invention

In order to overcome the MUR routing policy of the existing wireless sensor network based on 6LoWPAN because thereby information source to excessive excessive this drawback of energy consumption that causes of link hop count between the stay of two nights, the invention provides energy consumption, the reduction heavy degree of transitivity of packet and the wireless transducer network energy saving data transferring method based on 6LoWPAN of communication delay that a kind of effective reduction node transmits packet.

For the technical scheme that solves the problems of the technologies described above proposition is:

A wireless transducer network energy saving data transferring method based on 6LoWPAN, described wireless transducer network energy saving data transferring method comprises following process:

First,, by find an optimum interim assembled node TAN between information source and the stay of two nights, select the method for interim assembled node TAN as follows:

Step 1: with <A _i, A _j> represents connected node A _iwith node A _jlink, with <A ₀, A ₁..., A _n> represents with A ₀for information source node, A _nfor information destination node and through node A ₁, A ₂, A ₃... and A _n-1data package transmission path, altogether n jumps.

Step 2: calculate E by formula (19) _avg, and get Min_E=E (n) _avg, and remember Best_Pair={ n } (n).

$E_{\mathrm{avg}}\left(n\right)\&equiv;P_{\mathrm{Succ}}{E}_{\mathrm{Total}}^{\left(S\right)}+(1-P_{\mathrm{Succ}})E_{\mathrm{Total}}^{\left(F\right)}$

$=S_{\mathrm{IP}}\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{[1-{\left(q_i\right)}^N]}^M\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}+(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{[1-{\left(q_i\right)}^N]}^M)$

$\&CenterDot;\underset{t=1}{\overset{M}{\mathrm{\&Sigma;}}}t[\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}k\left(N\right[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_j^{\mathrm{\&gamma;}}]+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{N{q}_i}^{N+1}}{1-q_i}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}\left]\right)$

$\&CenterDot;{\left(q_j\right)}^N\underset{i=1}{\overset{j-1}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N]+(M-t)k\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}]$

$\&CenterDot;{(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N\left]\right)}^t{\left(\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}\right[1-{\left(q_i\right)}^N\left]\right)}^{M-t}---\left(19\right)$

Wherein, S _iPit is the bit number that IPv6 packet comprises; q _iand d _irespectively link <A _i-1, A _ithe packet loss of > and link <A _i-1, A _ithe distance of > two-end-point, i=1,2 ..., n; M represents the sheet number that IPv6 packet is divided into; N represents the maximum retransmission of MAC layer in one hop link; E _elecrepresent the power consumption of emitter and the every transmission of receiving circuit or recruiting unit's bit; ε _amprepresent that emission amplifier transmits at unit square rice the energy that every bit consumes; γ is path loss parameter;

Step 3: get n ₁=1;

Step 4: get n ₂=n-n ₁;

Step 5: calculate E by formula (19) _avg(n ₁) and E _avg(n ₂);

Step 6: if E _avg(n ₁)+E _avg(n ₂) <Min_E, get Min_E=E _avg(n ₁)+E _avg(n ₂), and remember Best_Pair={ n ₁, n ₂;

Step 7:n ₁=n ₁+ 1, if n ₁<n, forwards step 4 to;

Step 8: if Best_Pair has two value n ₁and n ₂, will be apart from information source n ₁the intermediate node of jumping is chosen for TAN, otherwise, between information source and the stay of two nights, do not establish TAN;

Then, after TAN chooses, information source node A ₀packet burst, be then that Mesh-under Routing method sends to TAN all bursts according to the MUR providing in 6LoWPAN agreement; The adaptation layer of the protocol stack of TAN is sent to network layer on all bursts that receive, and it is information source node A that all bursts are assembled into source IPv6 packet by network layer ₀the packet sending;

Afterwards, the network layer of TAN is extracted the object IP address of IPv6 packet head, searches routing table and obtains next-hop node address, then be routed to information destination node A by IPv6 packet burst and by MUR _n, then, information destination node is assembled into original IPv6 packet the data fragmentation receiving.

Technical conceive of the present invention is: by information source node A ₀with information destination node A _nbetween to choose an intermediate node be Temporary Assembling Node as " interim assembled node " TAN, make on the adaptation layer of TAN, the packet burst receiving to be submitted to network layer and carry out assembledly, and then the packet after assembled is carried out to burst and is forwarded to next-hop node.And on other intermediate node, packet burst is only submitted to adaptation layer and just carries out forward process.

With <A _i, A _j> represents connected node A _iwith node A _jlink, with <A ₀, A ₁.., A _n> represents with A ₀for information source node, A _nfor information destination node and through node A ₁, A ₂, A ₃... and A _n-1data package transmission path.At path <A ₀, A ₁..., A _n> is upper, selects certain intermediate node A _xas interim assembled node TAN.Like this, from source node A ₀to destination node A _nroute by interim assembled node A _xbe divided into two sections: L ₁and L ₂, wherein L ₁=<A ₀, A ₁..., A _x>, L ₂=<A _x, A _x+1..., A _n>.Refer to below about choosing of interim assembled node TAN.In the data transferring method that the technical program proposes, packet is when each node, and in interdependent node protocol stack, related protocol hierarchy as shown in Figure 2.

The people such as W.B.Heinzelman in October, 2002 at document 1: International Periodicals " IEEE Transactions on Wireless Communications " has been delivered paper " An application-specific protocol architecture for wireless microsensor networks " i.e. " a kind of application-specific protocol frame of wireless microsensor network " on " IEEE radio communication transactions " the 1st volume the 4th phase 660-670 page, has proposed following energy consumption model in this paper:

By the data packet transmission d rice distance of a k bit, the energy that transmit leg and recipient's radio circuit consume is respectively:

E _Tx(k,d)=k[E _elec+ε _ampd ^γ] （1）

E _Rx(k)=kE _elec （2）

Wherein, E _elecrepresent the power consumption of emitter and the every transmission of receiving circuit or recruiting unit's bit; ε _amprepresent that emission amplifier transmits at unit square rice the energy that every bit consumes; γ is path loss parameter, and its span is [2,4].Be " Heinzelman energy consumption model " hereinafter referred to as above-mentioned energy consumption model.Energy consumption when this technology adopts Heinzelman energy consumption model to weigh data transmission.

As everyone knows, the link of wireless sensor network is easy packet loss, insecure.For one from information source A ₀to stay of two nights A _ndata package transmission path <A ₀, A ₁..., A _n>, with q _irepresent link <A _i-1, A _ithe packet loss of >, p _i=1-q _ilink <A _i-1, A _i> transmits the success rate of packet, i=1, and 2 ..., n; Represent the sheet number that IPv6 packet is divided into M.In addition represent, the maximum retransmission of MAC layer in one hop link with N.That is to say, MAC layer is in the time transmitting same packet, if N continuous time is transmitted unsuccessfully, MAC layer just no longer continues to retransmit this packet.Therefore, for a burst, node A _i-1to next-hop node A _ithe probability of bust this is:

P _i ^(F)≡(q _i) ^N，i＝1，2，…，n (3)

So, node A _i-1to next-hop node A _ithe probability of transmission success is:

P _i ^(S)≡1-P _i ^(F)＝1-(q _i) ^N，i＝1，2，…，n (4)

Therefore,, under MUR routing policy, " an IPv6 packet that is divided into M burst is from information source A in chance event ₀send and by stay of two nights A _nassembled success " " M burst is all by stay of two nights A to be equivalent to chance event _nreceive ", its probability is:

$P_{\mathrm{Succ}}\&equiv;\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{\left[{P_i}^{\left(S\right)}\right]}^M=\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{[1-{\left(q_i\right)}^N]}^M\text{---}\left(5\right)$

Transmitting in successful situation, a burst is from node A _i-1to next-hop node A _iaverage transmission times be:

$N_i^S\&equiv;\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}j\left[{\left(q_i\right)}^{j-1}{p}_i\right]=p_i\frac{d}{d{p}_i}\left(\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}{q_i}^j\right)=p_i\frac{d}{d{p}_i}\left(\frac{q_i[1-{q_i}^N]}{1-q_i}\right)$

$N_i^S=\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}---\left(6\right)$

Obtained by formula (1), (2) and (6), transmitting in successful situation, the packet burst of a k bit is from node A _i-1to next-hop node A _iaverage energy consumption be:

$E_i^S\&equiv;k[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]\frac{1-(N+1){q_i}^N+{N{q}_i}^{N+1}}{1-q_i}---\left(7\right)$

Wherein, d _irepresent node A _i-1and A _ibetween distance.

Therefore, at information source A ₀a k bit sliced of sending can arrive stay of two nights A _nthis situation, path <A ₀, A ₁..., A _nthe total degree that the upper all nodes of > send packet is

$N^{\left(S\right)}\&equiv;\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{N}_i^S=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}---\left(8\right)$

Total energy consumption is

$E^{\left(S\right)}\&equiv;\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{E}_i^S=k\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}---\left(9\right)$

Can be obtained by formula (6), under MUR routing policy, if an IPv6 packet that is divided into M burst is successfully delivered to stay of two nights A _n, from information source A ₀send and by stay of two nights A _nassembled success, path <A ₀, A ₁..., A _nthe total degree that the upper all nodes of > send packet is

$N_{\mathrm{Total}}^{\left(S\right)}\&equiv;M\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{N}_i^S=M\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}---\left(10\right)$

Can be obtained by formula (7), if a bit number is S _iPiPv6 packet be successfully delivered to stay of two nights A _n, from information source A ₀send and by stay of two nights A _nassembled success, path <A ₀, A ₁..., A _nthe total energy consumption that the upper all nodes of > send packet is

$E_{\mathrm{Total}}^{\left(S\right)}\&equiv;S_{\mathrm{IP}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}---\left(11\right)$

Because a data slice is jumped at j the probability being dropped and is (j=1,2 ..., n), its corresponding transmission times is corresponding energy consumption is $N[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_j^{\mathrm{\&gamma;}}]+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}{N}_i^S[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}].$ Therefore, utilize formula (6) to obtain: at information source A ₀a k bit sliced of sending fails to arrive stay of two nights A _nthis situation, path <A ₀, A ₁..., A _nthe total degree that the upper all nodes of > send packet is

$N^{\left(F\right)}\&equiv;\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}(N+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}{N}_i^S)\left[P_j^{\left(F\right)}\underset{i=1}{\overset{j-1}{\mathrm{\&Pi;}}}{P}_i^{\left(S\right)}\right]$

$\left(12\right)$

$=\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}(N+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i})\left[P_j^{\left(F\right)}\underset{i=1}{\overset{j-1}{\mathrm{\&Pi;}}}{P}_i^{\left(S\right)}\right]$

Overall average energy consumption is

$E^{\left(F\right)}\&equiv;\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}k\left(N\right[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_j^{\mathrm{\&gamma;}}]+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}{N}_i^S[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}\left]\right)\left[P_j^{\left(F\right)}\underset{i=1}{\overset{j-1}{\mathrm{\&Pi;}}}{P}_i^{\left(S\right)}\right]---\left(13\right)$

By formula (3), (4) and (12):

$N^{\left(F\right)}\&equiv;\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}[N+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}]{\left(q_j\right)}^N\underset{i=1}{\overset{j-1}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N]---\left(14\right)$

By formula (3), (4), (6) and (13):

$E^{\left(F\right)}=\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}k\left(N\right[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_j^{\mathrm{\&gamma;}}]+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}\left]\right)---\left(15\right)$

$\&CenterDot;{\left(q_j\right)}^N\underset{i=1}{\overset{j-1}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N]$

With P ^(S)represent that a burst success is from information source A ₀be passed to stay of two nights A _nprobability.Can be obtained by formula (4): $P^{\left(S\right)}=\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{P}_i^{\left(S\right)}=\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N].$ Like this, in M burst, there is t burst to fail to arrive stay of two nights A _nprobability be [1-P ^(S)] t[P ^(S)] ^m-t, now, path <A ₀, A ₁..., A _nthe total degree that the upper all nodes of > send packet is tN ^(F)+ (M-t) N ^(S).So, obtained by formula (8) and (14): fail to be successfully delivered to stay of two nights A at an IPv6 packet that is divided into M burst _ncondition under, path <A ₀, A ₁..., A _nthe total degree that the upper all nodes of > send packet is

$N_{\mathrm{Total}}^{\left(F\right)}\&equiv;\underset{t=1}{\overset{M}{\mathrm{\&Sigma;}}}[t{N}^{\left(F\right)}+(M-t)N^{\left(S\right)}]{[1-P^{\left(S\right)}]}^t{\left[P^{\left(S\right)}\right]}^{M-t}$

$=\underset{t=1}{\overset{M}{\mathrm{\&Sigma;}}}[t{N}^{\left(F\right)}+(M-t)N^{\left(S\right)}]{(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N\left]\right)}^t{\left(\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}\right[1-{\left(q_i\right)}^N\left]\right)}^{M-t}$

$=\underset{t=1}{\overset{M}{\mathrm{\&Sigma;}}}\left[t\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}\right[N+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}\left]{\left(q_j\right)}^N\underset{i=1}{\overset{j-1}{\mathrm{\&Pi;}}}\right[1-{\left(q_i\right)}^N]---\left(16\right)$

$+(M-t)\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}]{(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N\left]\right)}^t{\left(\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}\right[1-{\left(q_i\right)}^N\left]\right)}^{M-t}$

Obtained by formula (6), (9) and (15): be S at a bit number _iPiPv6 packet fail to be successfully delivered to stay of two nights A _ncondition under, path <A ₀, A ₁..., A _nthe overall average energy consumption that the upper all nodes of > send packet is

$E_{\mathrm{Total}}^{\left(F\right)}=\underset{t=1}{\overset{M}{\mathrm{\&Sigma;}}}[t{E}^{\left(F\right)}+(M-t)E^{\left(S\right)}]{[1-P^{\left(S\right)}]}^t{\left[P^{\left(S\right)}\right]}^{M-t}$

$=\underset{t=1}{\overset{M}{\mathrm{\&Sigma;}}}t[\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}k\left(N\right[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_j^{\mathrm{\&gamma;}}]+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}\left]\right)$

$\left(17\right)$

$\&CenterDot;{\left(q_i\right)}^N\underset{i=1}{\overset{j-1}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N]+(M-t)k\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}]$

$\&CenterDot;{(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N\left]\right)}^t{\left(\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}\right[1-{\left(q_i\right)}^N\left]\right)}^{M-t}$

Wherein,

Obtained by formula (5), (10) and (16): under MUR routing policy, an IPv6 packet is from information source A ₀to stay of two nights A _naverage degree of transitivity be

$N_{\mathrm{avg}}\left(n\right)\&equiv;P_{\mathrm{Succ}}{N}_{\mathrm{Total}}^{\left(S\right)}+(1-P_{\mathrm{Succ}})N_{\mathrm{Total}}^{\left(F\right)}$

$=M\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{[1-{\left(q_i\right)}^N]}^M\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}+(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{[1-{\left(q_i\right)}^N]}^M)$

$\&CenterDot;\underset{t=1}{\overset{M}{\mathrm{\&Sigma;}}}\left[t\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}\right[N+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}\left]{\left(q_j\right)}^N\underset{i=1}{\overset{j-1}{\mathrm{\&Pi;}}}\right[1-{\left(q_i\right)}^N]$

$+(M-t)\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}]{(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N\left]\right)}^t{\left(\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}\right[1-{\left(q_i\right)}^N\left]\right)}^{M-t}$

Obtained by formula (5), (11) and (17): under MUR routing policy, an IPv6 packet is from information source A ₀to stay of two nights A _naverage energy consumption be

$E_{\mathrm{avg}}\left(n\right)\&equiv;P_{\mathrm{Succ}}{E}_{\mathrm{Total}}^{\left(S\right)}+(1-P_{\mathrm{Succ}})E_{\mathrm{Total}}^{\left(F\right)}$

$=S_{\mathrm{IP}}\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{[1-{\left(q_i\right)}^N]}^M\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}+(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{[1-{\left(q_i\right)}^N]}^M)$

$\&CenterDot;\underset{t=1}{\overset{M}{\mathrm{\&Sigma;}}}t[\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}k\left(N\right[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_j^{\mathrm{\&gamma;}}]+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{N{q}_i}^{N+1}}{1-q_i}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}\left]\right)$

$\&CenterDot;{\left(q_j\right)}^N\underset{i=1}{\overset{j-1}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N]+(M-t)k\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}]$

$\&CenterDot;{(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N\left]\right)}^t{\left(\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}\right[1-{\left(q_i\right)}^N\left]\right)}^{M-t}---\left(19\right).$

Beneficial effect of the present invention is mainly manifested in: can reduce the energy consumption of node transmission packet, number of retransmissions and the communication delay of packet.

Brief description of the drawings

Fig. 1 is the path examples figure that wireless sensor node forms;

Fig. 2 is that interim assembled node of the present invention is the protocol stack level figure that TAN relates to, and wherein, 1 represents physical layer, and 2 represent MAC layer, and 3 represent adaptation layer, and 4 represent network layer, and 5 represent transport layer, and 6 represent application layer;

Fig. 3 is the energy consumption comparison figure of method of the present invention and MUR.

Embodiment

Below in conjunction with accompanying drawing, the present invention will be further described.

With reference to Fig. 2 and Fig. 3, a kind of wireless transducer network energy saving data transferring method based on 6LoWPAN, described wireless transducer network energy saving data transferring method comprises following process:

First, between information source and the stay of two nights, find an optimum interim assembled node TAN by " TAN finding method ", select the method for interim assembled node TAN as follows:

Step 1: with <A _i, A _j> represents connected node A _iwith node A _jlink, with <A ₀, A ₁..., A _n> represents with A ₀for information source node, A _nfor information destination node and through node A ₁, A ₂, A ₃... and A _n-1data package transmission path, altogether n jumps.

Step 2: calculate E by formula (19) _avg, and get Min_E=E (n) _avg, and remember Best_Pair={ n } (n).

$E_{\mathrm{avg}}\left(n\right)\&equiv;P_{\mathrm{Succ}}{E}_{\mathrm{Total}}^{\left(S\right)}+(1-P_{\mathrm{Succ}})E_{\mathrm{Total}}^{\left(F\right)}$

$=S_{\mathrm{IP}}\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{[1-{\left(q_i\right)}^N]}^M\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}+(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{[1-{\left(q_i\right)}^N]}^M)$

$\&CenterDot;\underset{t=1}{\overset{M}{\mathrm{\&Sigma;}}}t[\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}k\left(N\right[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_j^{\mathrm{\&gamma;}}]+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{N{q}_i}^{N+1}}{1-q_i}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}\left]\right)$

$\&CenterDot;{\left(q_j\right)}^N\underset{i=1}{\overset{j-1}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N]+(M-t)k\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}]$

$\&CenterDot;{(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N\left]\right)}^t{\left(\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}\right[1-{\left(q_i\right)}^N\left]\right)}^{M-t}---\left(19\right)$

Wherein, S _iPit is the bit number that IPv6 packet comprises; q _iand d _irespectively link <A _i-1, A _ithe packet loss of > and link <A _i-1, A _ithe distance of > two-end-point, i=1,2 ..., n; M represents the sheet number that IPv6 packet is divided into; N represents the maximum retransmission of MAC layer in one hop link, that is to say, MAC layer is in the time transmitting same packet, if N continuous time is transmitted unsuccessfully, MAC layer just no longer continues to retransmit this packet; E _elecrepresent the power consumption of emitter and the every transmission of receiving circuit or recruiting unit's bit; ε _amprepresent that emission amplifier transmits at unit square rice the energy that every bit consumes; γ is path loss parameter, and its span is [2,4];

Step 3: get n ₁=1;

Step 4: get n ₂=n-n ₁;

Step 5: calculate E by formula (19) _avg(n ₁) and E _avg(n ₂);

Step 6: if E _avg(n ₁)+E _avg(n ₂) <Min_E, get Min_E=E _avg(n ₁)+E _avg(n ₂), and remember Best_Pair={ n ₁, n ₂;

Step 7:n ₁=n ₁+ 1, if n ₁<n, forwards step 4 to;

Step 8: if Best_Pair has two value n ₁and n ₂, will be apart from information source n ₁the intermediate node of jumping is chosen for TAN, otherwise, between information source and the stay of two nights, do not establish TAN;

Then, after TAN chooses, information source node A ₀packet burst, be then that Mesh-under Routing method sends to TAN all bursts according to the MUR providing in 6LoWPAN agreement; The adaptation layer of the protocol stack of TAN is sent to network layer on all bursts that receive, and it is information source node A that all bursts are assembled into source IPv6 packet by network layer ₀the packet sending;

Afterwards, the network layer of TAN is extracted the object IP address of IPv6 packet head, searches routing table and obtains next-hop node address, then be routed to information destination node A by IPv6 packet burst and by MUR _n, then, information destination node is assembled into original IPv6 packet the data fragmentation receiving.

Key of the present invention is above-mentioned " TAN finding method ", wherein, calculates E _avg(n) be that formula (19) is the core of TAN.In formula (19), there is following parameter: S _iPit is the bit number that IPv6 packet comprises; q _iand d _ibe link <A _i-1, A _ithe packet loss of > and link <A _i-1, A _ithe distance of > two-end-point, i=1,2 ..., n; M is the sheet number that IPv6 packet is divided into; N is the maximum retransmission of MAC layer in one hop link; E _elecit is the power consumption of the every transmission of emitter and receiving circuit or recruiting unit's bit; ε _ampbe that emission amplifier transmits at unit square rice the energy that every bit consumes; γ is path loss parameter.

In above-mentioned parameter, SIP is determined by IPv6 packet, M is set to S _iPthe smallest positive integral upper bound of/127 ratios, N is set to the value of attribute macMaxFrameRetries in IEEE 802.15.4 standard, reference literature 2:IEEE Computer Society. " IEEE 802.15.4 Standard for Wireless Medium Access Control (MAC) and Physical Layer (PHY) Specifications for Low-Rate Wireless Personal Area Networks (WPANs) ", 2006, it is IEEE computer society, " IEEE 802.15.4 marks the network medium access control of accurate – – low rate wireless personal area and physical layer specification ", 2006.Node is after sending packet, and the ETX that calculates the wireless link that uses is Expected Transmission Count, then gets q _ifor link <A _i-1, A _ithe inverse of the ETX of >; d _iutilize location technology or GPS to determine.In addition, the same with document [1], get E _elec=50 to receive joule every bit be nJ/bit; When distance is greater than 87 meters between node, get ε _ampthe every square meter of the every bit of=10 skin joule is pJ/bit/m ², and get γ=2; When distance is not more than 87 meters between node, get ε _amp=0.0013pJ/bit/m ⁴, and get γ=4.

Provide an example below so that energy-saving effect of the present invention to be described.Equate to be q at the packet loss of each link ₁=q ₂=...=q _nunder the condition of=q, formula (18) deformability is:

$N_{\mathrm{avg}}\left(n\right)=M{[1-q^N]}^{\mathrm{nM}}\frac{n[1-(N+1)q^N+{\mathrm{Nq}}^{N+1}]}{1-q}+(1-{[1-q^N]}^{\mathrm{nM}})$

$\&CenterDot;\underset{t=1}{\overset{M}{\mathrm{\&Sigma;}}}[t\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}(N+\frac{(j-1)[1-(N+1)q^N+{\mathrm{Nq}}^{N+1}]}{1-q})q^N{(1-q^N)}^{j-1}---\left(20\right)$

$+\frac{(M-t)n[1-(N+1)q^N+{\mathrm{Nq}}^{N+1}]}{1-q}]{{(1-{[1-q^N]}^n)}^t[1-q^N]}^{n(M-t)}$

Meanwhile, formula (19) deformability is:

$E_{\mathrm{avg}}\left(n\right)=S_{\mathrm{IP}}{[1-q^N]}^{\mathrm{nM}}\frac{[1-(N+1)q^N+{\mathrm{Nq}}^{N+1}]}{1-q}[{2\mathrm{nE}}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]+(1-{[1-q^N]}^{\mathrm{nM}})$

$\&CenterDot;\underset{t=1}{\overset{M}{\mathrm{\&Sigma;}}}t[\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}k\left(N\right[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_j^{\mathrm{\&gamma;}}]+\frac{[1-(N+1)q^N+{\mathrm{Nq}}^{N+1}]}{1-q}[2(j-1)E_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}{d}_i^{\mathrm{\&gamma;}}\left]\right)q^N{[1-q^N]}^{j-1}$

$+\frac{(M-t)k[1-(N+1)q^N+{\mathrm{Nq}}^{N+1}]}{1-q}[{2\mathrm{nE}}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{d}_i^{\mathrm{\&gamma;}}]]{(1-{[1-q^N]}^n)}^t{[1-q^N]}^{n(M-t)}---\left(21\right)$

Get E _elec=50 to receive joule every bit be nJ/bit, and get ε _ampthe every square meter of the every bit of=10 skin joule is pJ/bit/m ².In addition, get γ=2, d=15, M=10, N=3, and allow q=0.3, and 0.4,0.5. gets n=20, and information source and stay of two nights distance are 20 jumpings.Given " the TAN finding method " according to the present invention, in the time of q=0.4 and q=0.5, the node of jumping apart from information source 10 is chosen for TAN, and in the time of q=0.3, intermediate node is not all chosen for TAN, result of calculation is as shown in Figure 3.As can be seen from Figure 3: method of the present invention is lower than the energy consumption of MUR, and along with the increase of link failure rate q, method of the present invention is more obvious than the Energy Intensity Reduction of MUR.In the time that link failure rate q and n get other value, we can obtain similar conclusion.

The technology of the present invention can be widely used in the wireless sensor network based on IEEE 802.15.4 standard, makes sensor node transmit IPv6 packet with low energy consumption.Because wireless sensor network based on IEEE 802.15.4 standard uses, to exempt from license be that the industrial science medical treatment frequency range ISM of License-free is Industrial, Scientific and Medical frequency range communicates, along with the propelling of Internet of Things, the wireless sensor network based on IEEE 802.15.4 standard will be more and more universal.Therefore, the present invention has broad application prospects.

Claims (1)

1. the wireless transducer network energy saving data transferring method based on 6LoWPAN, is characterized in that: described wireless transducer network energy saving data transferring method comprises following process:

First,, by find an optimum interim assembled node TAN between information source and the stay of two nights, select the method for interim assembled node TAN as follows:

Step 1: with <A _i, A _j> represents connected node A _iwith node A _jlink, with <A ₀, A ₁..., A _n> represents with A ₀for information source node, A _nfor information destination node and through node A ₁, A ₂, A ₃... and A _n-1data package transmission path, altogether n jumps;

Step 2: calculate E by formula (19) _avg, and get Min_E=E (n) _avg, and remember Best_Pair={n} (n);

$\begin{array}{c}{E}_{\mathrm{avg}}\left(n\right)\&equiv;P_{\mathrm{Succ}}{E}_{\mathrm{Total}}^{\left(S\right)}+(1-P_{\mathrm{Succ}})E_{\mathrm{Total}}^{\left(F\right)}\\ =S_{\mathrm{IP}}\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{[1-{\left(q_i\right)}^N]}^M\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[2E_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]\frac{1-(N+1){q_i}^N+{q_i}^{N+1}}{1-q_i}+(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}{[1-{\left(q_i\right)}^N]}^M)\\ \&CenterDot;\underset{t=1}{\overset{M}{\mathrm{\&Sigma;}}}t[\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}k\left(N\right[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_j^{\mathrm{\&gamma;}}]+\underset{i=1}{\overset{j-1}{\mathrm{\&Sigma;}}}\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}[{2E}_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}\left]\right)\\ \&CenterDot;{\left(q_j\right)}^N\underset{i=1}{\overset{j-1}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N]+(M-t)k\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}[2E_{\mathrm{elec}}+{\mathrm{\&epsiv;}}_{\mathrm{amp}}{d}_i^{\mathrm{\&gamma;}}]\frac{1-(N+1){q_i}^N+{{\mathrm{Nq}}_i}^{N+1}}{1-q_i}\\ \&CenterDot;{(1-\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}[1-{\left(q_i\right)}^N\left]\right)}^t{\left(\underset{i=1}{\overset{n}{\mathrm{\&Pi;}}}\right[1-{\left(q_i\right)}^N\left]\right)}^{M-t}\end{array}---\left(19\right)$

Wherein, S _iPit is the bit number that IPv6 packet comprises; q _iand d _irespectively link <A _i-1, A _ithe packet loss of > and link <A _i-1, A _ithe distance of > two-end-point, i=1,2 ..., n; M represents the sheet number that IPv6 packet is divided into; N represents the maximum retransmission of MAC layer in one hop link; E _elecrepresent the power consumption of emitter and the every transmission of receiving circuit or recruiting unit's bit; ε _amprepresent that emission amplifier transmits at unit square rice the energy that every bit consumes; γ is path loss parameter; P _succrepresent M the probability that burst is all received by the stay of two nights; be illustrated in path <A under the condition that IPv6 packet is successfully delivered to the stay of two nights ₀, A ₁..., A _nthe upper all nodes of > send the total energy consumption of packet; be illustrated in IPv6 packet and fail to be successfully delivered under the condition of the stay of two nights, path <A ₀, A ₁..., A _nthe upper all nodes of > send the total energy consumption of packet; K represents a bit number that packet is shared; T is summation variable, and it is value t=1 successively, 2 ..., M;

Step 3: get n ₁=1;

Step 4: get n ₂=n-n ₁;

Step 5: calculate E by formula (19) _avg(n ₁) and E _avg(n ₂);

Step 6: if E _avg(n ₁)+E _avg(n ₂) <Min_E, get Min_E=E _avg(n ₁)+E _avg(n ₂), and remember Best_Pair={n ₁, n ₂;

Step 7:n ₁=n ₁+ 1, if n ₁<n, forwards step 4 to;

Step 8: if Best_Pair has two value n ₁and n ₂, will be apart from information source n ₁the intermediate node of jumping is chosen for TAN, otherwise, between information source and the stay of two nights, do not establish TAN;

Then, after TAN chooses, information source node A ₀packet burst, be then that Mesh-under Routing method sends to TAN all bursts according to the MUR providing in 6LoWPAN agreement; The adaptation layer of the protocol stack of TAN is sent to network layer on all bursts that receive, and it is information source node A that all bursts are assembled into source IPv6 packet by network layer ₀the packet sending;

Afterwards, the network layer of TAN is extracted the object IP address of IPv6 packet head, searches routing table and obtains next-hop node address, then be routed to information destination node A by IPv6 packet burst and by MUR _n, then, information destination node is assembled into original IPv6 packet the data fragmentation receiving.