JP3790245B2 - Communication module and communication method for wireless sensor network system - Google Patents

Description

  The present invention relates to a communication technique for performing data communication between nodes in a wireless sensor network system including a sensor node and a network node.

  In recent years, research and development of wireless sensor networks have progressed and have reached the level of practical use (for example, see Non-Patent Documents 1 and 2). The wireless sensor network is a network in which a small sensor node including various sensors such as a temperature sensor and an optical sensor and a network node that is an interface of a wide area network such as the Internet are wirelessly coupled.

  FIG. 13 is a diagram showing an example of the configuration of a wireless sensor network (see Non-Patent Documents 2 and 3). In FIG. 13, the wireless sensor network is composed of ten sensor nodes 101 and one network node 102. The network node 102 is connected to the host computer 103 by cable. The host computer 103 is connected to the Internet 104 by cable. Each sensor node 101 and network node 102 are wirelessly coupled. Each sensor node 101 transmits and receives data by performing wireless communication with the sensor node 101 in the vicinity thereof. Finally, data is collected in the network node 102. The network node 102 can transmit the collected data to the outside (the Internet 104) through the host computer 103.

  Currently existing sensor nodes include, for example, RSC WINS & Hida, Sensoria WINS (Non-Patent Document 4), UCLA's iBadge (Non-Patent Document 5), UCLA's Medusa MK-II (Non-Patent Document 6), Berkeley's Motes (Non-Patent Document 6) Patent Document 7), Berkeley Piconodes (Non-Patent Document 2), MIT's μAMPs (Non-Patent Document 8), and the like are known.

  FIG. 14 is a perspective view illustrating an example of a sensor node, and FIG. 15 is a diagram illustrating an example of a hardware architecture of the sensor node (see Non-Patent Documents 1 and 9). The sensor node 101 is made very small so as not to require installation space. The power source is a battery (or solar battery) 110 so that it can be freely installed in the environment. Accordingly, a DC-DC converter 111 for converting the applied voltage of the battery 110 into a desired voltage is provided. The sensor node 101 includes a sensor 112, an analog / digital converter (hereinafter referred to as “ADC”) 113, a memory 114, a micro controller unit (hereinafter referred to as “MCU”) 115, and an RF. A transceiver (RF transceiver) 116 and an antenna 117 are provided. The sensor 112 is various sensors such as a temperature sensor and an optical sensor. The analog signal generated in the sensor 112 is converted into digital data by the ADC 113 and input to the MCU 115. Then, the digital data is transmitted from the MCU 115 to the RF transceiver 116 and transmitted from the antenna 117 as an RF signal. The MCU 115 performs communication control, path generation control between modules, and power supply control. Control by the MCU 115 is executed by a reactive system operating system (OS) 118 stored in the memory 114.

  The sensor node 101 is required to have requirements such as real-time characteristics, hardware scale restrictions, low power consumption, and the like, and the reactive system OS 118 is adapted to these requirements. For example, TinyOS is known as such a reactive system OS 118 (see Non-Patent Document 9).

In general, a DC balanced modulation code (dc-balanced code) is selected as an encoding method for communication between the sensor nodes 101 so that the modulation signal is not affected by a DC component. This is because the encoded data modulates a VCO (Voltage Controlled Oscillator) during communication, and if a DC component remains, a center frequency tuning error occurs. In TinyOS, Manchester code or 4B6B code (see, for example, Patent Document 1 and Non-Patent Document 11) can be selected and used when performing weak power wireless communication between the sensor nodes 101 (Non-patent Document 11). Reference 10). These codes are DC balanced modulation codes. Manchester codes are encoded by transitions rather than information levels. Therefore, the bit rate is half the baud rate. On the other hand, in the case of 4B6B code, the bit rate is 2/3 of the baud rate.
US Pat. No. 4,824,463 Deborah Estrin, Akbar Sayeed, and Mani Srivastava, and Mani Srivastava, "Mobicom 2002 Tutorial T5: Wireless Sensor Networks", USA, ACM International Conference on Mobile Computing and Networking (MobiCom), September 2002 J.Rabaey, M.Ammer, J.Silva jr., D.Patel, and S.Roundy, "Picoradio supports ad hoc ultra-low power wireless networking", In IEEE Computer Magazine, USA, July 2000, pp. 42- 48. Philip Buonadonna, Jason Hill, and David Culler, "Active Message Communication for Tiny Networked Sensors", Submitted to Infocom USA, 2001 GJ Pottie, WJ Kaiser, "Wireless Integrated Network Sensors", Communications of the ACM, USA, May 2000, Vol. 43, No. 5, pp.551-8. Ivo Locher, "iBadge", [online], Networked & Embedded Systems Laboratory, University of California, [retrieved on November 20, 2003], Internet, <URL: http://nesl.ee.ucla.edu/projects / ibadge / ＞ Andreas Savvides and Mani B. Srivastava, "A Distributed Computation Platform for Wireless Embedded Sensing", Invited paper, to appear in the proceedings of ICCD 2002, Freiburg, Germany J. Hill, R. Szewczyk, A. Woo, S. Hollar, D. Culler, and K. Pister, "System architecture directions for network sensors", ASPLOS 2000, USA R. Min, M. Bhardwaj, et al., "An architecture for a power-aware distributed micro sensor node", in Proc. IEEE Workshop Signal Processing Systems (SiPS '00), USA, October 2000, pp. 581 -590. Jason Hill, Robert Szewczyk, Alec Woo, Seth Hollar, David Culler, Kristofer Pister, "System Architecture Directions for Networked Sensors", ACM SIGARCH Computer Architecture News, USA, December 2000, Volume 28, Issue 5, pp.93- 104, ACM Press New York, NY, USA Jerry Zhao and Ramesh Govindan, "Understanding Packet Delivery Performance In Dense Wireless Sensor Networks", The First ACM Conference on Embedded Networked Sensor Systems, Los Angeles, USA, November 2003 "ASH Transceiver Software Designer's Guide", [online], August 7, 2002, RF Monolithics Inc., [November 20, 2003 search], Internet, <URL: http://www.rfm.com/ corp / apnotes.htm>

  When data communication is performed between the sensor nodes, for example, when a 4B6B code is used, an 8-bit digital data packet to be transmitted is first composed of 4-bit data as shown in FIG. Divide into two data blocks. These are converted into 6-bit data blocks according to the conversion table. These 6-bit data blocks are sequentially transmitted.

  However, since data communication between sensor nodes is weak power RF communication, it is extremely susceptible to noise and fading from the power supply and surrounding devices. For this reason, bit errors frequently occur in wireless communication. If one data block is dropped due to a bit error, the error propagates to the packet transmitted following that data block, resulting in a burst error. That is, since an 8-bit data block is divided into two 4-bit data blocks and sequentially transmitted, if one data block is dropped, there is a deviation in the combination of data blocks that compose a packet. Because this deviation propagates, a burst error occurs.

  In order to prevent the occurrence of such an error, an error correction code is used in many communication systems. However, when an error correction code is used, the circuit scale for realizing it becomes large. Further, the more complicated the arithmetic processing, the more the real-time performance of data processing at the sensor node is impaired. In particular, sensor nodes need to communicate with as low power as possible using hardware resources of a limited scale. For this reason, it is desirable to avoid encoding involving complicated arithmetic processing as much as possible.

  On the other hand, in the wireless sensor network, most of the data transmitted / received by the sensor node is data obtained by digitizing the output from the sensor of each sensor node. Such data is not necessarily required to transmit and receive all data accurately, and some omission of data is allowed. This is because this kind of data is mostly used to grasp the general tendency (envelope tendency) of the physical quantity detected by the sensor. Accordingly, there is a situation in wireless sensor networks that there are few requests for restoring lost data using error correction codes.

  Accordingly, an object of the present invention is to provide a wireless sensor that can perform high-speed encoding / decoding processing without causing error propagation, and can be realized with low power consumption with limited hardware resources. It is to provide a communication technology for a network system.

  A first configuration of a communication module of a wireless sensor network system according to the present invention includes a sensor interface connected to a sensor, a transmission / reception control unit, and one or more sensor nodes including a wireless transmission / reception unit, Two types of one or more network nodes including a wireless transmission / reception unit that performs wireless communication with a sensor node, a transmission / reception control unit, and a network interface unit that exchanges data with an external network A communication module used for each node in a wireless sensor network system having a node, wherein the transmission / reception control unit converts a transmission / reception data block into front data (tail data) and tail data (tail data). Data block dividing means for dividing the data into two data blocks and outputting the data blocks; and the data A control code for adding a control code for identifying whether the data block is front data or tail data to the head of each data block output by the block dividing means And an adding means.

  Serial communication is generally used as a method for data transfer in wireless communication. The data length of the data transferred at that time is basically 1 byte (8 bits). In general, analog data is acquired from a sensor, ADC is applied, and converted digital data is generally handled with 1 byte as the minimum data length. Therefore, DC balanced modulation was adopted when 8bit digital data acquired from the sensor was transferred wirelessly. In addition, in order to satisfy the condition that the transfer with the minimum 1-byte data is performed, the original 8-bit data is divided into 4 bits and encoded to 6 bits to discriminate front data and tail data. By adding a 2-bit control code, 8bit data was obtained. As a result, it can be handled in the same manner as general serial transfer. In addition, the range of application is expanded. In addition, a simple communication method can be used to maintain real-time characteristics. On the contrary, it becomes easy to apply a complicated protocol. Furthermore, unlike a normal serial transfer, the addition of a control code can solve the problem of data loss that becomes a bottleneck in wireless communication.

  Also, even if one data block is dropped due to the occurrence of a code error in wireless communication, the data block is changed to the front data by referring to the control code of the data block following the dropped data block. Or tail data can be identified. Therefore, the original packet can be accurately restored even from the data block following the lost data block. Therefore, error propagation is prevented. Further, the arithmetic processing of the control code adding means only adds a control code, and does not require complicated arithmetic processing. Further, the calculation processing on the receiving side is only the reference of the control code and the assembly of the packet by that, and does not require complicated calculation processing. Therefore, it is possible to configure with a small amount of hardware resources, and it is possible to perform arithmetic processing with high speed and low power consumption. Therefore, the conditions required for the communication module used in the node of the wireless sensor network can be satisfied.

  According to a second configuration of the communication module of the wireless sensor network system according to the present invention, in the first configuration, the transmission / reception control unit is configured to transmit data of front data or tail data transmitted from another node. When the block is received by the wireless transmission / reception unit, control code detection means for detecting a control code added to the data block, and the control block for the data block B1 received by the wireless transmission / reception unit The control code C1 detected by the code detection means is compared with the control code C0 of the data block B0 received by the wireless transmission / reception unit immediately before the data block B1. If the code identifies the data, When the code code C1 is a code other than the code specifying the tail data, the data block B0 is deleted, and when the control code C0 is the code specifying the tail data, the control code C1 is the front code. An error block deleting means for performing data processing for deleting a data block B2 received immediately after the data block B1 when the code is a code other than a code for specifying data; .

  According to this configuration, when an error occurs in wireless communication between nodes and a data block is dropped, if the data block received immediately before the dropped data block is front data, the data block The first half of the packet composed of blocks is missing. Also, when the data block received immediately after the dropped data block is tail data, the latter half of the packet constituted by the data block is lost.

  Therefore, the error block deletion means includes the control code C1 detected by the control code detection means for the data block B1 received by the transmission / reception control section, and the transmission / reception control section immediately before the data block B1. The control code C0 of the received data block B0 is compared. When the control code C0 is a code specifying front data and the control code C1 is a code other than a code specifying tail data, the data block B0 is deleted. Further, when the control code C0 is a code specifying tail data and the control code C1 is a code other than a code specifying front data, the data block received immediately after the data block B1 Delete B2. As a result, if the latter half or the first half of a packet composed of data blocks received immediately before or after the missing data block is missing, the packet is deleted and an error packet is generated. Is prevented. Further, it is possible to prevent a burst error from occurring due to propagation of a shift in a combination of data blocks constituting a packet even for a data block received subsequent to the dropped data block.

  In addition, since a control code is added to data having the minimum data length and an error is determined for each piece of data on the receiving side, it is possible to minimize data discarded when an error is detected.

  According to a third configuration of the communication module of the wireless sensor network system according to the present invention, in the first or second configuration, the transmission / reception control unit is configured to transmit the front data and tail output from the data block dividing unit. Data block encoding means for encoding each of the two data blocks of data into a DC-balanced modulation code (DC-balanced code) is provided, and the control code adding means is encoded by the data block encoding means. A control code for identifying whether the data block is front data or tail data is added to the head of each of the converted data blocks.

  Since the data block encoding means encodes each data block into a DC balanced modulation code, the transmitted data has high DC balanced performance. Here, DC balance means that the number of 0s and 1s is almost equal in all periods of length N. In this way, by improving the DC balance performance of the data to be transmitted, no signal is lost when the information extracted outside by the high-pass filter is inserted into the original signal. In addition, since AC coupling can be employed in the amplifier at the reception stage of the wireless transceiver, the design of the automatic gain control circuit is simplified.

  In addition, data encoding is indispensable for improving DC balance performance during wireless communication. As described above, in order to maintain the minimum data transfer unit of 1 byte, the data is divided and recombined on the receiving side, so that a control code for discriminating front data and tail data is required. That is, by adopting the third configuration, the wireless communication circuit can be simplified and the data transfer can be simplified, and real-time communication and power saving can be achieved with a simple circuit.

  According to a fourth configuration of the communication module of the wireless sensor network system according to the present invention, in the third configuration, the data block dividing means is an 8-bit output from the sensor interface or the wireless transmission / reception unit. The transmission / reception data block is divided into two data blocks of 4-bit front data and 4-bit tail data, and the data block encoding means outputs each data of 4 bits. The block is encoded into a 6-bit DC modulation code.

  In this way, by adopting 4B6B encoding, the remaining 2 bits of 1 byte, which is the minimum data length when processing normal digital data, can be used as a control code, so it is possible to divide while ensuring DC balance performance It is also possible to implement error determination of the data that has been processed, and to perform efficient data transfer with a simple method.

  According to a fifth configuration of the communication module of the wireless sensor network system according to the present invention, in the third or fourth configuration, the transmission / reception control unit includes a code sequence of an input data block and a DC balanced modulation for the code sequence. A coding table in which a look-up table representing a correspondence relationship with codes is stored, and the data block coding means includes two data of front data and tail data output from the data block dividing means Each block is encoded into a DC balanced modulation code by referring to the encoding table.

  As described above, encoding is performed by a lookup table, thereby facilitating encoding processing. In addition, since the codeword registered in the lookup table in advance is used for encoding, the codeword used for encoding can be easily changed. In particular, when the source of noise that affects communication between nodes is a specific device or the like, the error rate for a specific codeword often increases. When UHP band RF communication is used, these noises depend on the surrounding environment (terrain, buildings), the presence of equipment that is the source of noise, and the location of the node. Codewords that are prone to error vary depending on the environment. Therefore, in such a case, depending on the location and environment, by registering the codeword used for encoding in the lookup table so as to avoid codewords that are likely to cause errors for each node. The influence of noise on communication between nodes can be reduced.

  Also, instead of reducing the influence of noise by performing complicated processing such as error checking and data retransmission / correction during communication, the influence of specific noise can be reduced by changing the lookup table. Can do. In addition, by adopting a method of storing as an encoding table, it becomes easy to change the table, and it becomes possible to respond to changes in the surrounding environment that affect wireless communication, and to stabilize communication. .

  According to a sixth configuration of the communication module of the wireless sensor network system according to the present invention, in the fifth configuration, the transmission / reception control unit assigns all code sequences usable in the DC balanced modulation code to a predetermined code sequence. When the pilot signal transmitting means transmits the pilot signal as a pilot signal in order and the pilot signal transmitted from the other node by the wireless transmitter / receiver, an error is detected for each DC balanced modulation code received. Pilot signal inspection means for detecting the occurrence rate or the number of error occurrences, and among the DC balanced modulation codes, the ones required in order from the one with the low error occurrence rate detected by the pilot signal inspection means or the one with the smallest number of error occurrences. Code selecting means for selecting a number of DC balanced modulation codes and registering them in the encoding table; For example, characterized in that is.

  As described above, the communication state at the time of data communication can be simplified by performing the inspection of the communication state by the pilot signal before the actual communication and establishing the best communication state. In local networks such as sensor networks, table reference agreements are made independently of other network protocols. Depending on the table reference agreement method, data from other networks It is also expected to protect the confidentiality.

  A first configuration of a communication method of a wireless sensor network system according to the present invention includes a sensor interface connected to a sensor, a transmission / reception control unit that performs transmission / reception control of data, and transmission / reception of data to / from other nodes wirelessly. One or a plurality of sensor nodes having a wireless transmission / reception unit for performing data transmission, a wireless transmission / reception unit for performing wireless communication with the sensor node, a transmission / reception control unit for performing transmission / reception control of data, and data transmission to an external network In a wireless sensor network system having two types of nodes, one or a plurality of network nodes provided with a network interface unit for sending and receiving, a communication method for performing data communication between the nodes. Block the transmit / receive data block into two data blocks: front data and tail data. A data block dividing procedure for dividing the data block, and at the head of each data block generated in the data block dividing procedure, to identify whether the data block is front data or tail data And a control code adding procedure for adding the control code.

  According to a second configuration of the communication method of the wireless sensor network system according to the present invention, in the first configuration, the data block of the front data or the tail data transmitted from another node is transmitted to the wireless transmission / reception unit. Is detected by the control code detection procedure for detecting the control code added to the data block and the control code detection procedure for the data block B1 received by the wireless transmission / reception unit. The control code C1 is compared with the control code C0 of the data block B0 received by the wireless transmission / reception unit immediately before the data block B1, and the control code C0 is a code for specifying the front data. In some cases, control code C1 The data block B0 is deleted, and when the control code C0 is a code specifying tail data, the control code C1 is a code other than a code specifying front data. If it is a code, it has an error block deletion procedure for performing data processing for deleting the data block B2 received immediately after the data block B1.

  According to a third configuration of the wireless sensor network system communication method of the present invention, in the first or second configuration, two data of front data and tail data generated by the data block dividing procedure are used. A data block encoding procedure for encoding each block into a DC balanced modulation code, and in the control code addition procedure, the head of each data block encoded in the data block encoding procedure And a process of adding a control code for identifying whether the data block is front data or tail data.

  According to a fourth configuration of the wireless sensor network system communication method of the present invention, in the third configuration, in the data block division procedure, 8 bits output from the sensor interface or the wireless transmission / reception unit. The transmission / reception data block is divided into two data blocks of 4-bit front data and 4-bit tail data. In the data block encoding procedure, each 4-bit data block is divided into 6 bits. It encodes to the direct current | flow modulation code of this.

  According to a fifth configuration of the communication method of the wireless sensor network system according to the present invention, in the third or fourth configuration, the transmission / reception control unit includes a code sequence of an input data block and a DC balanced modulation for the code sequence. A coding table storing a look-up table representing a correspondence relationship with codes, and in the data block coding procedure, the front data and tail data generated by the data block dividing procedure are stored. Each of the two data blocks is encoded into a DC balanced modulation code by referring to the encoding table.

The sixth configuration of the communication method of the wireless sensor network system according to the present invention is such that, in the fifth configuration, all of the DC balanced modulation codes that can be used before starting data communication between the nodes. Each of the received DC signals when the radio transmission / reception unit receives a pilot signal transmitted from another node as a pilot signal in a predetermined order from the radio transmission / reception unit. A pilot signal inspection procedure for detecting an error occurrence rate or the number of error occurrences for each balanced modulation code, and one of the DC balanced modulation codes having a low error occurrence rate detected by the pilot signal inspection procedure or the number of error occurrences The code that selects the necessary number of DC balanced modulation codes and registers them in the coding table in order from the smallest number And 択手 order,
It is characterized by having.

  As described above, according to the present invention, even when a data block transmitted in wireless communication between nodes is dropped, only a packet that is incomplete due to dropping is deleted, and another packet is deleted. It can be restored accurately. Therefore, error propagation can be prevented. Further, it can be configured with a small amount of hardware resources, and high-speed encoding / decoding processing is possible. In addition, it is possible to perform arithmetic processing with low power consumption using a limited battery power source such as a sensor node. Therefore, it is possible to provide a communication technique that satisfies the conditions required for the communication module used in the node of the wireless sensor network.

  The best mode for carrying out the present invention will be described below with reference to the drawings.

  FIG. 1 is a diagram illustrating the overall configuration of a wireless sensor network system according to a first embodiment of the present invention. The wireless sensor network system according to this embodiment includes a network node 1, a sensor node 2, and a host computer 3. Although only one network node 1 is shown in FIG. 1, there may be a plurality of network nodes. One or more sensor nodes 2 exist for one network node 1. Each sensor node 2 and network node 1 are connected by wireless communication. The network node 1 is connected to the host computer 3 by a cable. The host computer 3 is connected to the Internet 4. As a result, the network node 1 can be connected to the Internet 4 via the host computer 3.

  FIG. 1 shows a configuration in which a plurality of sensor nodes 2 are connected to one network node 1 at the same level (this configuration is called a “flat structure”). The connection between the network node 1 and the sensor node 2 may have a hierarchical structure as shown in FIG.

  Each sensor node 2 includes a sensor 5, an analog / digital signal processing device 6, an RF transceiver (RF transceiver) 7, and an antenna 7a (see FIG. 2). As the sensor 5, various sensors such as an optical sensor, a temperature sensor, and a sound sensor are used. The sensor 5 detects physical quantities such as light, temperature, and sound and outputs them as analog electrical signals. The analog / digital signal processing device 6 converts an analog electric signal output from the sensor 5 into a digital value, and between the nodes (referred to as the sensor node 2 and the network node 1, hereinafter the same) for the digital signal. Control communication with. The RF transceiver 7 is a device that actually performs RF wireless communication between the nodes.

  FIG. 2 is a block diagram showing the hardware configuration of the network node and the sensor node. The network node 1 is composed of three units: a base board 10, a PC board 11, and a battery 12. The sensor node 2 is composed of three units: a base board 10, a sensor board 13, and a battery 12. A common base board 10 is used for the network node 1 and the sensor node 2. In this way, by sharing parts between the network node 1 and the sensor node 2, each node can be manufactured at low cost.

  The base board 10 is equipped with a circuit having a function of performing wireless communication between nodes and a function of performing power supply control. Therefore, the base board 10 functions as a communication module and power supply control module for each node. A circuit for performing communication between the network node 1 and the host computer 3 is mounted on the PC board 11. The sensor board 13 includes a circuit that processes an analog signal output from the sensor 5 built in the sensor node 2 and converts the analog signal into a digital signal, and a circuit that controls power supplied to the sensor 5. The battery 12 supplies power to the base board 10, the PC board 11, and the sensor board 13.

  The base board 10 includes a central processing unit (hereinafter referred to as “CPU”) 15, a static random access memory (hereinafter referred to as “SRAM”) 16, and an RF transceiver 7. , A clock 17 and a power controller 18. The CPU 15 controls the entire circuit of the base board 10 and controls data communication with the PC board 11 or the sensor board 13. The SRAM 16 is provided to temporarily hold data generated during the arithmetic processing of the CPU 15. The RF transceiver 7 is the same as that described in FIG. The clock 17 generates a reference clock for performing a synchronous operation between the base board 10 and the PC board 11 or the sensor board 13. Further, the power control device 18 performs control to cut off or reduce the power supplied to the modules that are not operating in order to suppress the average power consumption of the entire node.

  In order to flexibly support various sensors used as the sensor 5 and to share parts, the PIC (Programmable Interrupt Controller), PLD (Programmable Logic Device), CPLD (Complex Programmable Logic Device) are used instead of the CPU 15. ), Programmable devices such as FPGA (Field Programmable Gate Array) can also be used.

  An RS232C driver 19 is mounted on the PC board 11 as a circuit for performing communication with the host computer 3. The RS232C driver 19 controls serial communication between the host computer 3 and the CPU 15 in accordance with a standard defined in RS232C (Recommended Standard 232 C).

  The sensor board 13 includes a low-pass filter (hereinafter referred to as “LPF”) 20, an analog-digital converter (hereinafter referred to as “ADC”) 21, and a power control device. 22 is provided. The LPF 20 cuts the DC component by removing the low frequency component of the analog signal input from the sensor 5. The ADC 21 quantizes the analog signal from which the low frequency is removed by the LPF 20 into a digital signal. A digital signal obtained by this quantization is output to the base board 10. The power control device 22 performs control to cut off or reduce the power supplied to the module that is not operating in order to suppress the average power consumption of the entire sensor board 13. That is, when the sensor 5 is not performing a sensing operation, the power of the other modules of the sensor board 13 is cut off or the low power consumption mode is set.

  FIG. 3 is a block diagram showing a functional configuration of the communication module in the sensor node 2 according to the first embodiment of the present invention. Since the functional configuration of the communication module of the network node 1 is the same as that in FIG. 3, only the functional configuration of the communication module of the sensor node 2 will be described here.

  The communication module of the sensor node 2 includes a packet generating unit 31, a data block dividing unit 32, a data block encoding unit 33, a control code adding unit 34, an encoding table 35, a control code detecting unit 36, an error The block deletion unit 37, the data block decoding unit 38, the packet assembly unit 39, the central control unit 40, the pilot signal transmission unit 41, the pilot signal inspection unit 42, and the code selection unit 43 are included. These functions are realized by the cooperation of the CPU 15 and the SRAM 16 shown in FIG. In FIG. 3, the sensor 5, the RF transceiver 7, the antenna 7a, the LPF 20, and the ADC 21 are the same as those in FIGS.

  The packet generator 31 reconstructs the sensing data output from the sensor 5 and quantized by the ADC 21 as an 8-bit packet. The data block dividing means 32 divides an input 8-bit packet into two 4-bit data blocks of front data and tail data, and outputs them.

  The data block encoding unit 33 encodes each of the two data blocks of the front data and the tail data output from the data block dividing unit 32 into a DC balanced modulation code. As the DC balanced modulation code, Manchester code (PE (Phase Encoding) code), mBnB code, mBnT code, nBIC code, or the like can be used. In this embodiment, 4B6B code is adopted. This is because 2 bits of control code are added to 6 bits of data encoded by 4B6B encoding to form 1 byte of data.

  Here, the input when the data block encoding means 33 performs encoding and the correspondence relationship for each code are stored in the encoding table 35. The encoding table 35 is a lookup table. The data block encoding means 33 encodes a 4-bit input data block at high speed with reference to the encoding table 35, and outputs a 6-bit encoded data block. FIG. 4 shows an example of a code conversion table of data blocks.

  The control code adding means 34 adds a 2-bit control code to the head of the 6-bit encoded data block output from the data block encoding means 33. The control code is a code for identifying whether the data block is front data or tail data. When the data block is front data, a control code “01” is attached, and when the data block is tail data, a control code “10” is attached. From the viewpoint of improving the data rate, the control code should be as short as possible. On the other hand, it is necessary to realize DC balance with only the control code, and to ensure that DC balance can be maintained even when added to data. Therefore, a 2-bit control code is used in this embodiment.

  When the RF transceiver 7 receives a data block of front data or tail data transmitted from another node, the control code detection means 36 detects a control code added to the data block.

  The error block deletion means 37 includes the control code C1 detected by the control code detection means 36 for the data block B1 received by the RF transceiver 7, and the RF transceiver 7 immediately before the data block B1. The received data block B0 is compared with the control code C0 to detect an error occurring in the data block during communication. That is, when the control code C0 is a code specifying the front data and the control code C1 is a code other than the code specifying the tail data, the error block deleting means 37 is a data block. Delete B0. Further, when the control code C0 is a code specifying tail data and the control code C1 is a code other than a code specifying front data, the data block received immediately after the data block B1 Delete B2. Then, the error block deleting means 37 outputs a 6-bit data block in which the first 2-bit control code is deleted for the data block not to be deleted.

  The data block decoding unit 38 decodes the 6-bit data block output from the error block deletion unit 37 by 4B6B conversion to generate a 4-bit decoded data block. Here, the data block decoding means 38 performs a decoding process with reference to the encoding table 35, when decoding.

  The packet assembling means 39 combines the decoded front data block and tail data block, and restores the original 8-bit packet. The restored packet is output to the central control unit 40.

  The central control unit 40 controls the flow of data packets. When the data output from the sensor 5 is input from the packet generating unit 31, the data is output to the data block dividing unit 32. Further, when a packet transferred from another node is input from the packet assembling means 39, it is output to the data block dividing means 32. (In the case of the network node 1, when the packet transferred from the other node is input from the packet assembling means 39, the central control unit 40 sends it to the RS232C driver of the PC board 11. Output.)

  The pilot signal transmission unit 41, the pilot signal inspection unit 42, and the code selection unit 43 are modules for reducing the influence of specific noise generated between the nodes as a stage before starting communication of sensing data between the nodes. It is.

  The pilot signal transmitting means 41 outputs all code sequences usable in the DC balanced modulation code to the control code adding means 34 as pilot signals in a predetermined order. That is, all codes usable for communication are transmitted as pilot signals to other nodes via the RF transceiver 7. When the RF transceiver 7 receives a pilot signal transmitted from another node, the pilot signal checking unit 42 detects an error occurrence rate or the number of error occurrences for each received DC balanced modulation code. The code selection means 43 selects the necessary number of DC balanced modulation codes in order from the one with the lowest error rate detected by the pilot signal checking means 42 or the smallest number of error occurrences among the DC balanced modulation codes. Are registered in the encoding table 35.

  In this way, between the nodes, using the pilot signal, the code word that minimizes the error rate is selected as the code word used for communication and stored in the encoding table 35. As a result, encoding is performed while avoiding the use of codewords that are prone to errors, depending on the location and environment of the node. Therefore, it is possible to suppress an increase in error rate depending on the environment (terrain, building) around the node and the presence of equipment that is a source of noise.

  The details of the code word selection operation using the pilot signal will be described later.

  The operation and experimental example of the wireless sensor network system according to the first embodiment configured as described above will be described below.

[1] Selection of Code Word First, details of the code word selection operation described above will be described. FIG. 5 is a flowchart showing the flow of the code word selection operation. The code word selection operation is performed by cooperation of two nodes (sensor node 2 or network node 1). Hereinafter, the nodes that perform the code word selection operation are denoted as N1 and N2. A node that transmits a test code (pilot symbol) is referred to as a transmitting node N1, and a node that receives a test code is referred to as a receiving node N2.

  First, the central control unit 40 of the transmission node N1 transmits a wakeup message to the reception node N2 by the RF transceiver 7 (S1). In the receiving node N <b> 2, when the wake-up message is received by the RF transceiver 7, the central control unit 40 outputs a wake-up mode switching signal to the power control device 18. In response to this, the power control device 18 starts supplying power to each module related to the code word selection operation in the base board 10 of the receiving node N2, and enters the wakeup mode (S21).

  Next, the central control unit 40 of the transmission node N1 transmits a test mode message (test mode message) to the reception node N2 by the RF transceiver 7 (S2). In the receiving node N2, when the test mode message is received by the RF transceiver 7 (S22), the pilot signal checking means 42 initially sets the error counter {e (k)} to 0 for each codeword to be subjected to the communication test. (S23). Here, the error counter is a variable for counting the number of errors that occur when communicating a code word for a code word for which a communication test is performed. In the present embodiment, since the 4B6B code is used as the DC balanced modulation code, it is possible to select and use 16 codewords from 20 codewords as shown in FIG. The error counter {e (k)} has a one-to-one correspondence with these 20 codewords.

  Next, the pilot signal transmission unit 41 of the transmission node N1 and the pilot signal inspection unit 42 of the reception node N2 set the variable i representing the number of tests and the variable j representing the test code index held as internal variables to 0. Initialization is performed (S3, S24). Here, the test code refers to a 6-bit code word for performing the 20 communication tests shown in FIG. Hereinafter, these code words are denoted as p (j) (j = 0, 1,..., 19).

  Next, in the transmission node N 1, the pilot signal transmission unit 41 outputs a test code p (j) to the control code addition unit 34. The control code adding means 34 adds a control code to the input test code p (j) and transmits it from the RF transceiver 7 to the receiving node N2 (S4). In the receiving node N2, when the test code p (j) is received by the RF transceiver 7 (S25), the pilot signal checking means 42 checks whether or not the correct test code p (j) is received (S26). . Here, it is determined in advance that the test codes are sequentially transmitted from the top to the bottom of the table shown in FIG. Therefore, the pilot signal inspection means 42 can determine whether or not the correct test code p (j) has been received by comparing the received test code with the code word determined in the order of FIG. . If the received test code is an error as a result of the determination, the value of the error counter e (j) is incremented by 1 (S27).

Next, the pilot signal transmission unit 41 of the transmission node N1 and the pilot signal inspection unit 42 of the reception node N2 increment the variable j by 1 (S5, S28). As a result, when the value of the variable j does not reach the maximum value N _c (= 20) (S6, S29), the process returns to the operation of steps S4 and S25, and the same test operation is performed for the next test code. repeat.

In steps S6 and S29, when the value of the variable j reaches the maximum value N _c (= 20), the pilot signal transmitting means 41 of the transmitting node N1 and the pilot signal checking means 42 of the receiving node N2 set the variable i to 1. Is incremented by 1 and variable j is initialized to 0 (S7, S30). If the value of the variable i has not reached the maximum value N _test (S8, S31), the process returns to the operations of steps S4 and S25, and the same test operation is repeated again from the 0th test code. Here, N _test represents the number of communication tests performed for all test codes. The value of N _test is set to an appropriate number of times that the error characteristic can be detected with sufficient accuracy.

In steps S8 and S31, when the value of the variable i reaches the maximum value _Ntest , the code selection means 43 of the receiving node N2 determines the error counter {e (k); k = 0,. , 19} are compared, and 16 test codes p (k) are selected in descending order of the value of e (k), and the 16 code words c (k) (k = 0, ..., 15) (S32). The code selection means 43 transmits the selected code word {c (k)} to the transmission node N1 by the RF transceiver 7 (S33). The code selection means 43 stores the received code word {c (k)} in the encoding table 35 (S34). On the other hand, in the transmission node N1, when the RF transceiver 7 receives the code word {c (k)} (S9), the code selection unit 43 stores the received code word {c (k)} in the encoding table 35. (S10).

  When the above operation is completed, the power control devices 18 of the transmission node N1 and the reception node N2 stop supplying power again and transition to a sleep mode (S11, S35).

  By the operation as described above, a code word for performing communication between the transmission node N1 and the reception mode N2 is selected and stored in the encoding table 35 of the transmission node N1 and the reception mode N2. In the subsequent communication between the transmission node N1 and the reception mode N2, communication is performed using the codeword stored in the encoding table 35.

[2] Sensing Data Communication Next, an operation when the sensor node 2 transfers sensing data will be described.

  FIG. 7 is a flowchart showing the overall flow of operation when the sensor node 2 transfers sensing data. The sensing data transfer operation is performed by cooperation of two nodes (sensor node 2 or network node 1). Hereinafter, the nodes that perform the sensing data transfer operation are denoted as N3 and N4. A node that requests sensing data is referred to as a data requesting node N3, and a node that transmits sensing data is referred to as a data transmitting node N4.

  First, the central control unit 40 of the data request node N3 transmits a wakeup message to the data transmission node N4 by the RF transceiver 7 (S41). In the data transmission node N4, when the wake-up message is received by the RF transceiver 7, the central control unit 40 outputs a wake-up mode switching signal to the power control device 18. In response to this, the power control device 18 starts supplying power to each module related to the sensing data transfer operation in the base board 10 of the data transmission node N4, and enters the wakeup mode (S51). ).

  Next, the central control unit 40 of the data request node N3 transmits a data request to the data transmission node N4 by the RF transceiver 7 (S42). In the data transmission node N4, when the data request is received (S52), the central control unit 40 transmits a start message (start message) to the data request node N3 by the RF transceiver 7 (S53). When the data request node N3 receives the start message, the data request node N3 transits to a reception mode (request mode) (S43).

  Subsequently, the data transmission node N4 transmits sensing data (S54), and the data request node N3 receives sensing data (S44). Details of the operations in steps S44 and S54 will be described later.

  Finally, when all the sensing data has been transferred, the central control unit 40 of the data transmission node N4 transmits a transmission end code EOD (End of Data) to the data request node N3 (S55), and enters the sleep mode. Transition to (sleep mode) (S56). When the data request node N3 receives the transmission end code EOD (S45), the data request node N3 transitions from the reception mode to the sleep mode (S46).

  FIG. 8 is a flowchart showing details of operations in steps S45 and S55 of FIG. 7, and FIG. 9 is a diagram showing processing in packet encoding.

  In the data transmission node N4, the central control unit 40 transmits a start message to the data request node N3 (S53), and then transmits a packet of sensing data to be transmitted to a packet generation unit (in the case of data relay, Obtained from the SRAM 16). Then, the data block dividing means 32 divides the acquired 8-bit packet into two 4-bit data blocks (front data B1 and tail data B2) (see S71, FIG. 9A). .

  Next, the data block encoding unit 33 refers to the encoding table 35 and performs 4B6B encoding processing on the two data blocks B1 and B2 obtained by the division (S72, FIG. 9B). )reference). Then, the control code adding means 34 adds a control code “01” to the head of the data block for the front data B1, and for the tail data B2, the control code “01”. A control code “10” is added to the head (S73, see FIG. 9B). Since each control code is a DC balanced code, even if the control code is added, the DC balance of each data block is not lost.

  The two generated 8-bit data blocks are wirelessly transmitted from the RF transceiver 7 to the data request node N3 (S74, S75).

  The data transmission node N4 repeats the operations from the above steps S71 to S75 until the transfer of all sensing data is completed (S76). When all the sensing data has been transferred, the central control unit 40 transmits an EOD to the data requesting node N3 (S55), and proceeds to the next operation described above.

  On the other hand, in the data request node N3, when the RF transceiver 7 receives the data block B1 in the reception mode (S61), the control code detection means 36 obtains the control code of the first 2 bits of the data block B1. It is detected and it is checked whether or not it is “01” (S62). If the control code is not "01", the error block deleting means 37 deletes the data block and waits for the next data block B1 (S61).

  On the other hand, when the control code is “01” in step S62, the data block decoding means 38 refers to the encoding table 35 and converts the data block B1 received by the 4B6B inverse transform to the 4-bit data. Decode to block B1 ′ Then, the packet assembling means 39 temporarily stores the decrypted data block B1 'in the internal register (S63).

  When the RF transceiver 7 receives the next data block B2 (S64), the control code detecting means 36 detects the control code of the first 2 bits of the data block B2, which is “10”. It is inspected (S65). When the control code is “10”, the data block decoding unit 38 refers to the encoding table 35 and decodes the data block B2 received by 4B6B inverse transform into a 4-bit data block B2 ′. To do. Then, the packet assembling means 39 combines the decoded data block B2 'and the data block B1' stored in the internal register, and restores the packet (S66). The central control unit 40 stores the restored packet in the SRAM 16 (or, when the data request node N3 is the network node 1, outputs it to the host computer 3 via the RS232C driver 19).

  On the other hand, if the control code is not “10” in step S65, the error block deleting means 37, if the control code is “01” (S67), the data stored in the internal register of the packet assembling means 39. Rewrite the block with the newly received data block (S68), and return to step S64. On the other hand, if the control code is not “01”, the error block deleting means 37 deletes the received data block B2 (S69) and returns to step S61.

  With the above operation, a data block in which an error has occurred is deleted, and a packet can be restored without causing a combinational shift.

  When transmission of all sensing data is completed, the RF transceiver 7 receives the EOD and shifts to the operation after step S45.

[3] Experimental Example Finally, the effect of the present invention was confirmed by an experiment using a prototype, and the result will be described below.

  FIG. 10 is an external view of the network node and the sensor node according to the first embodiment used in the experiment. (A) is the network node 1, and (b) is the sensor node 2. In this experiment, the wireless communication of the RF transceiver 7 used a frequency band of 303.1 MHz. The baud rate in wireless communication was 19.2 kbps. The base board 10 used a clock of 8 MHz. The output voltage of the battery was 3.5V, and the operating voltage Vcc of the base board 10, the PC board 11, and the sensor board 13 was 3.3V. Further, a helical antenna was used as the antenna 7a.

  FIG. 11 is a layout diagram of network nodes, sensor nodes, and peripheral devices used in the experiment. The experiment was performed indoors, and the distance between the network node 1 and the sensor node 2 was 50 cm. Instead of the sensor 5 input, a sine wave was input to the sensor node 2 side by a signal generator. The network node 1 is connected to the host computer 3 by an RS232C standard serial cable. The power supply of each node used 3.3V output AC adapter instead of the battery.

  Data transmitted from the sensor node 2 was received by the network node 1 and waveform data output to the host computer 3 was observed.

  FIG. 12 is a diagram showing received data waveforms obtained in the experiment. (A) is a waveform of reception data obtained when communication is performed without using a control code, and (b) is a reception data obtained when communication is performed by the communication method described in the first embodiment. It is a waveform of data. In (a), it is clear that noise is generated from the lost data blocks, and the observations are unsightly.

  In contrast, when communication is performed by the communication method described in the first embodiment, it is understood that noise is removed and clean sensing data is obtained as shown in (b).

It is a figure showing the whole structure of the wireless sensor network system which concerns on Example 1 of this invention. It is a block diagram showing the hardware constitutions of a network node and a sensor node. It is the block diagram which showed the module structure of the communication module in the sensor node which concerns on Example 1 of this invention. It is a figure which shows an example of the code conversion table of a data block. It is a flowchart showing the flow of selection operation | movement of a code word. It is a list of the code words which can be used with 4B6B encoding system. It is a flowchart showing the whole flow of operation | movement in case a sensor node transfers sensing data. It is a flowchart showing the detail of operation | movement of step S45, S55 of FIG. It is a figure showing the process in the encoding of a packet. It is an external view of the network node and sensor node which concern on Example 1 used in experiment. It is a layout view of network nodes and sensor nodes and peripheral devices used in the experiment. It is a figure which shows the received data waveform obtained by experiment. It is a figure which shows an example of a structure of a wireless sensor network. It is a perspective view showing an example of a sensor node. It is a figure which shows an example of a structure of the hardware architecture of a sensor node. It is a figure explaining the division | segmentation into the data block at the time of packet transmission.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Network node 2 Sensor node 3 Host computer 4 Internet 5 Sensor 6 Analog / digital signal processor 7 RF transceiver (high frequency radio transceiver)
7a Antenna 10 Base board 11 PC board 12 Battery 13 Sensor board 15 Central processing unit (CPU)
16 SRAM
17 Clock 18 Power control device 19 RS232C driver 20 Low-pass filter (LPF)
21 Analog-to-digital converter (ADC)
22 Power control device 31 Packet generation means 32 Data block division means 33 Data block coding means 34 Control code addition means 35 Encoding table 36 Control code detection means 37 Error block deletion means 38 Data block decoding means 39 Packet assembly means 40 Central controller 41 Pilot signal transmission means 42 Pilot signal inspection means 43 Code selection means

Claims (10)

One or a plurality of sensor nodes including a sensor interface connected to the sensor, a transmission / reception control unit, and a wireless transmission / reception unit;
One or a plurality of network nodes including a wireless transmission / reception unit that performs wireless communication with the sensor node, a transmission / reception control unit, and a network interface unit that exchanges data with an external network;
A communication module used for each node in a wireless sensor network system having two types of nodes,
The transmission / reception control unit includes:
A data block dividing means for dividing a transmission / reception data block into two data blocks of front data and tail data and outputting the data blocks;
Control code adding means for adding a control code for identifying whether the data block is front data or tail data to the head of each data block output by the data block dividing means When,
Control code detection means for detecting a control code added to the data block when the wireless transmission / reception unit receives a data block of front data or tail data transmitted from another node;
The control code C1 detected by the control code detection means for the data block B1 received by the wireless transmission / reception unit, and the data block received by the wireless transmission / reception unit immediately before the data block B1 The control code C0 of B0 is compared, and when the control code C0 is a code specifying front data and the control code C1 is a code other than a code specifying tail data, the data block B0 When the control code C0 is a code other than the code specifying the front data when the control code C0 is a code specifying the tail data, the data received immediately after the data block B1・ Delete block B2 And error block deletion means for performing data processing,
A communication module comprising: The transmission / reception control unit includes:
Data block encoding means for encoding each of the two data blocks of front data and tail data output from the data block dividing means into a DC balanced modulation code;
The control code adding means identifies at the head of each data block encoded by the data block encoding means whether the data block is front data or tail data. 2. The communication module according to claim 1 , wherein the control code is added. The data block dividing means converts an 8-bit transmission / reception data block output from the sensor interface or the wireless transmission / reception unit into two data blocks of 4-bit front data and 4-bit tail data. It is divided into blocks and output.
3. The communication module according to claim 2, wherein said data block encoding means encodes each 4-bit data block into a 6-bit DC modulation code. The transmission / reception control unit includes:
A coding table storing a look-up table representing a correspondence between a code sequence of an input data block and a DC balanced modulation code corresponding thereto;
The data block encoding means encodes each of the two data blocks of the front data and the tail data output from the data block dividing means into a DC balanced modulation code by referring to the encoding table. The communication module according to claim 2 or 3, wherein The transmission / reception control unit includes:
Pilot signal transmitting means for transmitting all code sequences usable in the DC balanced modulation code as pilot signals from the wireless transmission / reception unit in a predetermined order;
When the wireless transmission / reception unit receives a pilot signal transmitted from another node, pilot signal checking means for detecting an error occurrence rate or the number of error occurrences for each received DC balanced modulation code;
Of the DC balanced modulation codes, the coding table is selected by selecting the necessary number of DC balanced modulation codes in order from the one with the low error rate detected by the pilot signal checking means or the one with the few error occurrence numbers. Code selecting means to be registered in
The communication module according to claim 4, further comprising: One or more sensor nodes including a sensor interface connected to the sensor, a transmission / reception control unit that performs transmission / reception control of data, and a wireless transmission / reception unit that transmits / receives data to / from other nodes wirelessly;
One or more networks including a wireless transmission / reception unit that performs wireless communication with the sensor node, a transmission / reception control unit that performs transmission / reception control of data, and a network interface unit that transmits / receives data to / from an external network Nodes,
In a wireless sensor network system having two types of nodes, a communication method for performing data communication between nodes,
A data block division procedure for dividing a transmission / reception data block into two data blocks of front data and tail data;
Control code addition for adding a control code for identifying whether the data block is front data or tail data to the head of each data block generated in the data block dividing procedure Procedure and
A control code detection procedure for detecting a control code added to the data block when the wireless transmission / reception unit receives a data block of front data or tail data transmitted from another node;
The control code C1 detected by the control code detection procedure for the data block B1 received by the wireless transceiver, and the data block received by the wireless transceiver immediately before the data block B1 The control code C0 of B0 is compared. When the control code C0 is a code for specifying the front data and the control code C1 is a code other than the code for specifying the tail data, the data block B0 If the control code C0 is a code other than the code specifying the front data when the control code C0 is a code specifying the tail data, the data received immediately after the data block B1・ Deleting block B2 And error block deletion procedure for performing data processing,
A communication method characterized by comprising: A data block encoding procedure for encoding each of the two data blocks of the front data and the tail data generated by the data block division procedure into a DC balanced modulation code;
In the control code addition procedure, in order to identify whether the data block is front data or tail data at the head of each data block encoded by the data block encoding procedure. 6. The communication method according to claim 5 , wherein a process of adding the control code is performed. In the data block division procedure, an 8-bit transmission / reception data block output from the sensor interface or the wireless transmission / reception unit is converted into two data blocks of 4-bit front data and 4-bit tail data. Split into blocks,
8. The communication method according to claim 7, wherein in the data block encoding procedure, each 4-bit data block is encoded into a 6-bit DC modulation code. The transmission / reception control unit includes a coding table in which a lookup table representing a correspondence relationship between a code sequence of an input data block and a DC balanced modulation code corresponding thereto is stored,
In the data block encoding procedure, each of the two data blocks of the front data and the tail data generated in the data block dividing procedure is converted into a DC balanced modulation code by referring to the encoding table. 9. The communication method according to claim 7 , wherein encoding is performed. Prior to starting data communication between the nodes, a pilot signal transmission procedure for transmitting all code sequences usable in the DC balanced modulation code as pilot signals in a predetermined order from the radio transceiver unit;
When the wireless transceiver receives a pilot signal transmitted from another node, a pilot signal inspection procedure for detecting an error occurrence rate or the number of error occurrences for each received DC balanced modulation code;
Among the DC balanced modulation codes, a necessary number of DC balanced modulation codes are selected in order from the one with the low error occurrence rate detected in the pilot signal inspection procedure or the one with the few error occurrences, and the coding table A code selection procedure to be registered in
The communication method according to claim 9 , further comprising:

Images