CN103997403A - Embedded monitoring system for carrying out data transmission of wireless sensor based on compressed sensing (CS) - Google Patents
Embedded monitoring system for carrying out data transmission of wireless sensor based on compressed sensing (CS) Download PDFInfo
- Publication number
- CN103997403A CN103997403A CN201410201013.5A CN201410201013A CN103997403A CN 103997403 A CN103997403 A CN 103997403A CN 201410201013 A CN201410201013 A CN 201410201013A CN 103997403 A CN103997403 A CN 103997403A
- Authority
- CN
- China
- Prior art keywords
- data
- matrix
- node
- wireless sensor
- acceleration transducer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Mobile Radio Communication Systems (AREA)
Abstract
The invention provides an embedded algorithm for carrying out data transmission of a wireless sensor based on compressed sensing (CS). The algorithm has the advantages of being small in occupied memory and high in computation speed, and is particularly suitable for the intelligent wireless sensor with limited hardware resources. By means of the algorithm, data encryption and lost packet recovery based on CS are achieved on an Imote 2 intelligent wireless sensor platform, the unreliability of wireless data transmission can be effectively compensated for, and the requirement for rapid data collection performance in health monitoring of a structure can be met. Under support of the embedded algorithm, time-consuming and power-consuming data packet retransmission protocols of the wireless sensor can be replaced by faster and simpler transmission protocols, and incomplete data received by a base station can be recovered on a computer.
Description
(1) technical field
What patent of the present invention related to is the Embedded algorithm of the wireless sensor data transmission based on compressed sensing.It relates to wireless telecommunications, signal processing, civil engineering works structure health monitoring technical field.
(2) background technology
In current field of civil engineering, along with the construction of aging and more and more large and complex structures of a lot of building structure, more and more important role is being played the part of in health monitoring.Monitoring system not only can provide reference for the state estimation of structure, also can provide foundation for the maintenance of structure and maintenance.Its effect is directly connected to fail safe and the availability of structure.
The most basic in structure monitoring system is transducer and data acquisition system thereof, now widely used wired data acquisition method to have data acquisition reliable and stable, the advantage such as little affected by noise, but its loaded down with trivial details installation work need to spend larger cost.Therefore begin one's study in the world at present and use wireless senser and wireless data acquisition system.The advantage of wireless senser and radio sensing network thereof is not only can remove the engineering time to grow, invest large and loaded down with trivial details installation work, saves cost, and wireless senser intelligent processing capacity, can embed signal processing algorithm.In the time applying to monitor large bridge, the advantage of wireless senser is more outstanding.
But, because being easily subject to environmental interference, the transfer of data of wireless senser causes data packet loss, and the reliability of Wireless Data Transmission often needs special processing to be ensured.Wherein repeating transmission is conventional lost packet compensation method.But because communication environment on bridge is complicated and changeable, data packet loss is often unavoidable, repeats not only power consumption relatively consuming time of transmission, can not ensure in essence the complete reliability of data in the time that communication success rate is lower.In the time that incomplete data are used to carry out structure identification and assessment, the accuracy of its result is also affected.
On the other hand, the development of compressed sensing technology (CS) has brought the impact of the property changed on the collection of signal.By utilizing the sparse property of signal, compressed sensing technology can be to carry out the complete information of picked up signal much smaller than the sample frequency of sampling thheorem.This indirectly sampled form of recovering afterwards of first encrypting, can reduce the data volume that needs storage and transmission greatly; Be particularly useful for the intelligent wireless sensor of hardware resource scarcity.Utilize CS algorithm to be encrypted the data of Wireless Acceleration Sensor collection, the data packet loss in wireless transmission process can obtain effective compensation by later stage CS recovery algorithms, thereby improves the unreliability of Wireless Data Transmission.
But, CS cryptographic algorithm is embedded to the encrypted transmission but difficulties of intelligent wireless sensor platform for data.Its reason is that traditional C/S cryptographic algorithm depends on an accidental projection matrix data that gather are encrypted to conversion; And in this matrix, every bit need be produced at random by identical probability-distribution function.Use so full random matrix to be encrypted acceleration information, transducer must shift to an earlier date each element in storage matrix.In the time that acceleration information is longer, (establishing number of data points is n), stores this matrix (n
2individual floating type element) needed internal memory very large (as, work as n=1,, store with double-precision floating point ocra font ocr the static memory that this matrix needs 8MB at 000 o'clock); And ciphering process amount of calculation is large, consuming time longer.These problems make traditional CS scrambled matrix be confined to theoretical research, and do not have too large practical use.
Up to now, the typical intelligent wireless sensor platform that applies in the world monitoring structural health conditions field only has very limited datarams, computing capability and supply of electric power conventionally.Such as, the Mica of Berkeley University's research and development has 4KB Dram, the static memory of 128KB+512KB and the CPU frequency of 7.38MHz; The wireless senser node of professor Lynch of University of Michigan research and development has the Dram of 26KB+512KB, the static memory of 448KB and the CPU frequency of 40MHz; Professor Spencer of University of Illinois has the Dram of 256KB+32MB, the variable C PU dominant frequency of the static store of 32MB and 13-416MHz with the Imote2 intelligent wireless sensor platform of Intel R & D Cooperation.Clearly, on most intelligent wireless sensor, the storage of CS scrambled matrix and computations are difficult to realize.Even for the superior Imote2 of hardware device, embedding a larger CS scrambled matrix is huge burden equally.Therefore, a kind of embedded cryptographic algorithm of the limited CS of hardware resource that is suitable for is transmitted most important for the data wireless of realizing based on CS.The present invention, just for this problem, provides a kind of CS cryptographic algorithm that is suitable for embedding intelligent wireless sensor.This algorithm committed memory is little, and computational speed is fast, meets the hardware constraints of most of intelligent wireless sensor.Based on this algorithm, the present invention has realized data encryption and the loss recovery based on CS on Imote2 intelligent wireless sensor platform, unreliability that can effective compensation Wireless Data Transmission, and the rapid data that meets monitoring structural health conditions is collected performance requirement.Under the support of this Embedded algorithm, the packet of wireless senser power consumption relatively consuming time repeats host-host protocol and can be replaced by more quick succinct host-host protocol, and the deficiency of data that receive base station can be recovered on computers.
(3), summary of the invention
The object of the invention is to propose the Structural Vibration Monitoring System based on wireless sensor network of the data encryption of a kind of CS of comprising and the Embedded algorithm of transmission.。
Described in the present invention thes contents are as follows:
Core of the present invention is that one group of CS that can run on intelligent wireless sensor platform encrypts and transmission system.The CS scrambled matrix of its embedding is produced by random demodulation method (RD), and to only have arrowband, center be nonzero element to the scrambled matrix that the method produces, and nonzero element is all made up of+1/-1; If this matrix is Φ
rD, than traditional CS scrambled matrix Φ, wireless senser storage Φ
rDrequired memory headroom is little; And use Φ
rDfast to data computations speed, its implementation is as follows:
The first step: on the calculating kernel of intelligent wireless sensor, by embedding the scrambled matrix Φ of transducer
rD, it is the coded signal y that the initial data x of n converts equal length to that transducer is collected to length, i.e. y=Φ
rDx; Then data y is wirelessly transmitted to Data Collection base station by wireless sensor node;
Second step: in base station, the deficiency of data that data sink receives
wherein there is n
lindividual data point is lost in wireless transmission process; According to the order of Packet Generation, specified data packet loss quantity and position, thereby determine matrix
by Φ
rDget rid of corresponding obliterated data and put corresponding n
lrow element obtains, and its dimension is (n-n
l) × n;
The 3rd step: at base station computing terminal, set up data reconstruction model
in formula, α is Fourier coefficient; F is inverse discrete Fourier transform matrix; Adopt l
1minimized method solves,
and
obtain optimization Fourier coefficient
The 4th step: calculate deciphering reconstruct initial data afterwards
Effect of the present invention is as follows:
The present invention has realized at intelligent wireless sensor node and has calculated on kernel the data that gather are carried out to CS encryption and transmission; On base station, realize the recovery to imperfect enciphered data.Whole process has promoted the reliability of wireless monitor system transfer of data.
On Imote2 intelligent wireless sensor platform (as shown in Figure 2), the present invention can be encrypted simultaneously and transmit 3-axis acceleration signal, and is applicable to different sample frequencys, meets the demand of Monitoring of Structural Vibration.
The present invention can effectively compensate by the bridge monitoring acceleration signal below 20% data loss rate, reduces the error that data packet loss causes.
Brief description of the drawings
Wireless Data Transmission and the recovery of Fig. 1 based on compressed sensing
Fig. 2 embeds program flow diagram (flow process of Imote2 calculating and communication)
Fig. 3 application example 1
Fig. 4 application example 2
Fig. 5 application example 3
The time-domain signal error energy ratio of Fig. 6 reconstruct data and deficiency of data
The frequency-region signal error energy of Fig. 7 reconstruct data and deficiency of data is than embodiment:
Enforcement of the present invention is based on Imote2 intelligent wireless sensor platform and Illinois monitoring structural health conditions kit (http://shm.cs.uiuc.edu/).The operating system of Imote2 is TinyOS, and code language is nesC.
CS scrambled matrix generates and embeds with the form of code the memory of Imote2 in advance at outer computer.The RD method that this scrambled matrix is proposed by this patent produces, and its producing method is as follows:
1. produce a random sequence p that length is n (the enciphered data segment length that n is initial data, in the realization of this invention, n is 2000), its each element p
iobey the symmetrical Bernoulli Jacob P (p that distributes
i=± 1)=1/2.P is expressed as to a diagonal matrix P, its diagonal element P
ii=p
i.
2. produce a rolling average matrix H (except center tyre remainder is 0); Its bandwidth is determined by the sparse property of signal, in the realization of this invention, is chosen as 5.
3. produce CS scrambled matrix: Φ
rD=HP.Can find out, because H is sparse arrowband matrix, Φ
rDsparse arrowband matrix equally.Due to the existence of a large amount of 0 elements, use Φ
rDlittle to initial data computation degree, be suitable for the limited wireless senser of computing capability.
In fact, be sequence of random variables owing to only there being p, only p need be embedded in Imote2 static memory.In ciphering process, Φ
rDalso do not need by complete generation, only it need be realized on the calculating kernel of Imote2 the transformation of data, memory headroom and computational speed have further been saved in processing like this.
Gathered after acceleration information at Imote2, acceleration time-histories data are broken down into the data segment of 2000; The CS/RD scrambled matrix that every one piece of data is embedded Imote2 in advance converts the enciphered data section of 2000 to; Imote2 is spliced to form enciphered data section final enciphered data and is stored in static memory.When Imote2 node receives the data record order of base station, Imote2 is by enciphered data packing packing and send continuously packet to base station.Base station is receiving after packet, finds out unreceived part (packet loss) and with 0 replacement obliterated data; Final data are decoded and are recovered with CS to computer by base-station transmission.
The Imote2 wireless senser that embeds CS Data Encryption Transmission algorithm of the present invention is used to a series of bridge monitoring and acceleration collection experiment.These experimental verifications feasibility of the present invention and practicality.Below with three instantiations, effect of the present invention is described; It represents respectively situation common in three kinds of wireless telecommunications (without packet loss, random loss and continual data package dropout).Example 1:
Fig. 3 has shown that enciphered data y does not have the situation of packet loss in transmitting procedure.In this case,
ciphering process and restructuring procedure are completely reciprocal, and initial data is error free Exact recovery after reconstruct, as shown in Fig. 3 (e).
Example 2:
Fig. 4 has shown that the data of wireless senser in the time running into random loss recover example: when one section of acceleration information of encrypting with Embedded algorithm is when in transmitting procedure, random loss rate reaches 8.8%, and the data reconstruction effect of CS recovery algorithms.Can find out, the signal after recovery and primary signal error are less, have high similitude at frequency domain.
Example 3:
Fig. 5 has shown the data recovery example of wireless senser in the time running into continual data package dropout: this section of acceleration information packet loss in wireless transmission process reaches 16.3%, and nearly 5% continual data package dropout has occurred.Now, although reconstructed error is larger, the present invention appoints the right quality of data decline that dropout causes that improved.
The error analysis of example 2 and 3:
Fig. 6,7 provided data recover the signal of example 2 and 3 on time domain and frequency domain with error energy than (dB).It is defined as follows.Can find out, this reconstructing method has improved the energy Ratios of signal and error greatly, has obviously improved the quality of data that has packet loss.
Time-domain signal and error energy ratio:
Frequency-region signal and error energy ratio:
Wherein x and X are primary signal machine frequency spectrum;
and
for reconstruction signal and frequency spectrum thereof.
Claims (5)
1. one kind based on intelligent radio acceleration transducer and compressed sensing (Compressive Sensing, CS) the bridge structure vibration acceleration data Fast Acquisition System of algorithm, formed by following part: intelligent radio acceleration transducer node, CS encipheror, applicable wireless transmission protocol, computer data recovery routine; Wherein intelligent radio acceleration transducer node comprises: node acceleration transducer, node calculates kernel, node memory, node wireless transceiver and accessory circuit.
2. the structural vibration acceleration information Fast Acquisition System based on intelligent radio acceleration transducer and CS algorithm according to claim 1, is characterized in that: CS scrambled matrix is stored in the static memory of node; Gathered after tentation data at node acceleration transducer, node calculate kernel data are encrypted and by data storing in node holder; By the collection command triggers of base station, node utilizes fast and reliable host-host protocol to send enciphered data back to base station; The data that collect base station will be by computer program reconstruct and for structural analysis and diagnosis.
3. according to the bridge vibration information Fast Acquisition System based on intelligent radio acceleration transducer and CS algorithm described in claim 1-2, it is characterized in that: the CS scrambled matrix of its embedding is by random demodulation method (Random Demodulator, RD) produce, to only have arrowband, center be nonzero element to the scrambled matrix that the method produces, and nonzero element is all made up of+1/-1; If this matrix is Φ
rD, than traditional CS scrambled matrix Φ, wireless senser storage Φ
rDrequired memory headroom reduces greatly; And use Φ
rDfast to data computations speed, its implementation is as follows:
The first step: on the calculating kernel of intelligent wireless sensor, by embedding the scrambled matrix Φ of transducer
rD, it is the coded signal y that the initial data x of n converts equal length to that transducer is collected to length, i.e. y=Φ
rDx; Then data y is wirelessly transmitted to Data Collection base station by wireless sensor node;
Second step: in base station, the deficiency of data that data sink receives
wherein there is n
lindividual data point is lost in wireless transmission process; According to the order of Packet Generation, specified data packet loss quantity and position, thereby determine matrix
by Φ
rDget rid of corresponding obliterated data and put corresponding n
lrow element obtains, and its dimension is (n-n
l) × n;
The 3rd step: at base station computing terminal, set up data reconstruction model
in formula, α is Fourier coefficient; F is inverse discrete Fourier transform matrix; Adopt l
1minimized method solves,
and
obtain optimization Fourier coefficient
The 4th step: calculate deciphering reconstruct initial data afterwards
4. according to the bridge vibration information Fast Acquisition System based on intelligent radio acceleration transducer and CS algorithm described in claim 1-3, it is characterized in that: CS scrambled matrix generates and embed with the form of code the memory of Imote2 in advance at outer computer.The RD method that this scrambled matrix is proposed by this patent produces, and its producing method is as follows:
(1) produce a random sequence p that length is n (the enciphered data segment length that n is initial data), its each element pi obeys the symmetrical Bernoulli Jacob P (p that distributes
i=± 1)=1/2.P is expressed as to a diagonal matrix P, its diagonal element P
ii=p
i.
(2) produce a rolling average matrix H (except center tyre remainder is 0);
Its bandwidth is determined by the sparse property of signal.
(3) produce CS scrambled matrix: Φ
rD=HP.Can find out, because H is sparse arrowband matrix, Φ
rDsparse arrowband matrix equally; Due to the existence of a large amount of 0 elements, use Φ
rDlittle to initial data computation degree, be suitable for the limited wireless senser of computing capability.
5. the bridge vibration information Fast Acquisition System based on intelligent radio acceleration transducer and CS algorithm according to claim 1, it is characterized in that: CS scrambled matrix is generated by random demodulator (RD), it is non-zero element that the scrambled matrix that RD produces only has arrowband, center.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410201013.5A CN103997403A (en) | 2014-05-13 | 2014-05-13 | Embedded monitoring system for carrying out data transmission of wireless sensor based on compressed sensing (CS) |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410201013.5A CN103997403A (en) | 2014-05-13 | 2014-05-13 | Embedded monitoring system for carrying out data transmission of wireless sensor based on compressed sensing (CS) |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103997403A true CN103997403A (en) | 2014-08-20 |
Family
ID=51311414
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410201013.5A Pending CN103997403A (en) | 2014-05-13 | 2014-05-13 | Embedded monitoring system for carrying out data transmission of wireless sensor based on compressed sensing (CS) |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103997403A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106788885A (en) * | 2016-12-07 | 2017-05-31 | 上海交通大学 | A kind of compress ecg data rate Automatic adjusument wireless transmitting system and transmission method |
| CN109959511A (en) * | 2019-04-10 | 2019-07-02 | 浙江大学 | A kind of marine tidal-current energy machine unit real-time state monitoring method based on vibration signal |
| CN111887840A (en) * | 2020-08-28 | 2020-11-06 | 绍兴梅奥心磁医疗科技有限公司 | Whole body multipath electrocardio real-time wireless monitoring system and method |
| CN112859799A (en) * | 2019-11-26 | 2021-05-28 | 北京交通大学长三角研究院 | Method for compensating data packet loss in network control system based on compressed sensing |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070273620A1 (en) * | 2006-05-29 | 2007-11-29 | Sony Corporation | Image display |
| CN102970707A (en) * | 2012-12-20 | 2013-03-13 | 李惠 | Wireless transmission data loss recovery method on basis of compressed sensing |
-
2014
- 2014-05-13 CN CN201410201013.5A patent/CN103997403A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070273620A1 (en) * | 2006-05-29 | 2007-11-29 | Sony Corporation | Image display |
| CN102970707A (en) * | 2012-12-20 | 2013-03-13 | 李惠 | Wireless transmission data loss recovery method on basis of compressed sensing |
Non-Patent Citations (1)
| Title |
|---|
| 马坚伟: "压缩感知及其应用：从稀疏约束到低秩约束优化", 《信号处理》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106788885A (en) * | 2016-12-07 | 2017-05-31 | 上海交通大学 | A kind of compress ecg data rate Automatic adjusument wireless transmitting system and transmission method |
| CN106788885B (en) * | 2016-12-07 | 2021-01-01 | 上海交通大学 | Electrocardiosignal compression rate self-adaptive adjustment wireless transmission system and transmission method |
| CN109959511A (en) * | 2019-04-10 | 2019-07-02 | 浙江大学 | A kind of marine tidal-current energy machine unit real-time state monitoring method based on vibration signal |
| CN112859799A (en) * | 2019-11-26 | 2021-05-28 | 北京交通大学长三角研究院 | Method for compensating data packet loss in network control system based on compressed sensing |
| CN111887840A (en) * | 2020-08-28 | 2020-11-06 | 绍兴梅奥心磁医疗科技有限公司 | Whole body multipath electrocardio real-time wireless monitoring system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105181122A (en) | Mechanical vibration signal data compression and acquisition method | |
| CN103997403A (en) | Embedded monitoring system for carrying out data transmission of wireless sensor based on compressed sensing (CS) | |
| Jayawardhana et al. | Compressive sensing for efficient health monitoring and effective damage detection of structures | |
| CN103280084A (en) | Data acquisition method for multi-parameter real-time monitoring | |
| Li et al. | Compressed sensing in multi-hop large-scale wireless sensor networks based on routing topology tomography | |
| CN103763696B (en) | Energy-saving sensor data collection method based on cross layer safe compression | |
| CN102970707A (en) | Wireless transmission data loss recovery method on basis of compressed sensing | |
| US11589286B2 (en) | Systems and methods for compressed sensing in wireless sensor networks | |
| CN115459919A (en) | Internet of things universal terminal security authentication system, method and device and Internet of things communication terminal | |
| CN102355752A (en) | Wireless sensor network compressed sensing measurement matrix based on expander graph and reconfiguring method | |
| CN102546293B (en) | High speed network flow network address measuring method based on Hash bit string multiplexing | |
| Wielandt et al. | TDD LoRa and delta encoding in low-power networks of environmental sensor arrays for temperature and deformation monitoring | |
| CN103327530B (en) | A kind of data transmission method in wireless sensor network | |
| CN108495314A (en) | Based on the encrypted underwater wireless sensing network information transferring method of chaos compressed sensing | |
| CN105682141B (en) | Collection method for wireless sensor network data based on neighbours' auxiliary | |
| Kolo et al. | A simple data compression algorithm for wireless sensor networks | |
| Li et al. | Chaotic deep network for mobile D2D communication | |
| CN105530012A (en) | Compressed sensing based wavelet domain sparse one-dimensional oil well data compression and reconstruction method | |
| CN105979548A (en) | Time-space compression network coding based data transmission method in wireless sensor network | |
| CN107483541A (en) | A kind of online task immigration method based on rolling time horizon | |
| CN107171771A (en) | Monitoring structural health conditions wireless sensor data loses Distortionless Embedded algorithm | |
| CN102802199A (en) | Data processing method for wireless sensor real-time monitoring system based on compressive sensing | |
| Islam et al. | Structural health monitoring by payload compression in wireless sensors network: An algorithmic analysis | |
| Padalkar et al. | Energy efficient data gathering in wireless sensor networks and internet of things with compressive sensing at sensor node | |
| CN103297959A (en) | Method for encryption in sensor network, encryption device and sensor network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140820 |
|
| WD01 | Invention patent application deemed withdrawn after publication |