Current development board: TelosB gateway node + Tmote sky sensor node, with temperature, humidity and light intensity sensors.
The development board adopts the 6lowpan networking under the contiki operating system.
|
Origin of technology
|
UC Berkeley
|
|
CPU
|
8MHz TI MSP430, 10KB RAM
|
|
Communication chip
|
CC2420
|
|
communication waves
|
2.4GHz radio, 250Kbps high-speed data transmission
|
|
MAC protocol
|
IEEE 802.15.4 / ZigBee
|
|
external flash
|
1M Flash
|
|
antenna
|
Built-in, integrated in the circuit board
|
|
operating system
|
TinyOS 1.1 or higher
|
|
write program
|
Write via USB port
|
|
powered by
|
Two AA batteries (AA batteries)
|
|
Temperature and humidity sensing components
|
Sensirion Sht11
|
|
light sensing components
|
Hamamatsu S1087 (S1087-1)
|
|
LED
|
3, respectively 0(Red), 1(Green), 2(Blue)
|
1 Classification of sensor nodes:
Low-performance nodes: sensor nodes have relatively low computing capabilities, storage capabilities, and transmission capabilities, and low power consumption. They can only collect and transmit physical scalar information.
High-performance nodes: Sensor nodes have relatively powerful computing, storage, and transmission capabilities, but have high power consumption, and can collect and transmit video images and other information, and perform some local processing.
2 low performance platforms:
The main feature of a low-performance platform is that its processing capacity, storage capacity, and transmission capacity are greatly limited, and the platform has few resources available. The low-performance platforms commonly used in the market are as follows:
a. Mica series
The Mica series includes Mica, Mica2, MicaZ, IRIS and Cricket, all of which are provided by memsic company (formerly Crossbow company). The Mica series nodes all include an 8-bit AVR processor with a frequency of 4-16MHz and a 128KB programmable flash. The processing capabilities and storage capabilities of the Mica series nodes are relatively similar, but there are big differences in transmission channels and rates. The frequency of Mica node transmission channel includes 433MHz and 916MHz, and the rate is 40kbps. The frequency of Mica2 node transmission channel includes 315, 433, 868, 916MHz, and the rate is 40kbps. MicaZ and IRIS nodes use the IEEE802.15.4 standard, the frequency of the transmission channel is 2.4GHz, and the rate is 250kbps. Mica series platform storage capacity is very limited RAM 4-8KB, ROM 512KB. The Cricket node is an upgraded version of the Mica2 node. Based on the Mica2, an ultrasonic sending and receiving device is added, which can perform ultrasonic positioning. In addition, the Mica series platform is powered by two AAA batteries and provides I/O interfaces for users to connect peripherals. Compared with MICA products, the IRIS node platform has a longer operating distance, up to 500 meters outdoors, with a rate of 250kbps, and is an RF transceiver based on the IEEE802.15.4/ZIGBEE protocol. IRIS can cooperate with a high-performance platform to form a powerful WSN system, and IRIS is mainly responsible for data transmission. The Mica series is currently offered by MEMSIC as a commercial kit.
b. TelosB/Tmote Sky
TelosB and Tmote Sky node platforms are similar to the hardware architecture of Mica series platforms. TelosB nodes are developed by the University of California, Berkeley and licensed to Crossbow Company. The Tmote Sky node platform is provided by Sentilla Corporation (formerly Moteiv). TelosB and TmoteSky use 16-bit TI MSP430 microcontroller as microcontroller, with a main frequency of 8MHz and 10KB of RAM. In addition, TelosB and TmoteSky node platforms integrate some sensors, such as temperature, humidity, light sensors, etc., and use the USB interface to connect with the host, which is convenient for programming and debugging. 6-10 pins are provided for user to connect peripherals. In the market, these two kinds of nodes are sold. The TelosB node is provided by MEMSIC, and the Tmote Sky node is provided by Sentilla. TelosB and Tmote node platforms have relatively large user groups in academia.
c. Eyes/EyesIFX v2 The
Eyes node platform was developed by a three-year European project, and its architecture is similar to TelosB/TmoteSky. Using 16-bit microprocessor, 16KB RAM, 2KB ROM, transmission channel frequency 868MHz, the rate reaches 115.2kbps. In addition, the Eyes node platform integrates accelerometers, temperature, light, pressure sensors, and more. Use the RS232 serial port to connect with the host. EyesIFX v2 is developed by Infineon[4], and its performance is similar to Eyes node platform in all aspects. EyeslFX v2 adopts 16-bit TI MSP430 microcontroller, transmission channel 868MHz, and USB programming interface. In WSN application systems, low-performance platforms usually undertake simple sensing tasks. They are usually equipped with low-power processors to reduce power consumption and are relatively inexpensive. Therefore, in WSN applications, low-performance platforms are widely used.
Summary & Comparison: Low-performance platforms mainly focus on several parameters such as processor type, storage capacity, radio frequency chip, communication frequency, and operating system.
3 high-performance nodes:
In addition to sensing physical information, high-performance nodes should also have local processing functions, multi-hop communication functions, and large-scale data transmission functions, but these additional functions cannot be provided by low-performance platforms. Some advanced operations, such as network management, image and video information transmission, and image and video information processing require the platform to have a powerful processor, sufficient storage space, and sufficient transmission bandwidth. To meet these requirements, researchers developed a high-performance WSN platform. Currently there are mainly the following:
a. iSense
iSense is developed by coalesenses and provides a complete set of software and hardware platforms, with the goal of meeting the needs of both industry and academia [5]. The hardware platform of iSense is assembled on demand by modules with independent functions, including sensor module, core module, and power module. Users can purchase relevant modules to assemble and build a specific WSN system according to research or use needs. The iSense hardware platform is built around the core module iSense CoreModule3. CoreModule3 is equipped with a JN5148 processor [6], 32-bit RISC, main frequency 16MHz, 128KB RAM, 512KB ROM, power consumption, 6mA in working state, sleep 3uA in state, channel 2.4GHz, rate 250-667kbit/s. iSense provides different power supply modules, including specific modules in the solar power supply module, which are easily connected to the main control board to form an application system. iSense provides software modules that complement the hardware. The iSense software module integrates a series of "out of the box" services and protocols, including routing, time synchronization, wireless programming, etc. In addition to a proprietary mesh protocol stack, IPv4 and IPv6 protocol stacks are also provided. iSense also provides a SHAWN wireless sensor network simulation platform. iSense provides C++ API, and its development environment is its own software development system iSenseOS, which provides functions similar to operating systems, but does not support specific operating systems (such as linux, TinyOS, etc.). Typical application fields of iSense [7]: environmental monitoring and preventive protection (such as temperature and humidity monitoring in museums), factory automation monitoring and maintenance, supply chain and asset management monitoring, indoor security
monitoring (doors, windows, etc.). iSense is currently offered as a commercial kit by coalesenses.
b、Imote/Imote2
Imote and Imote2 are high-performance wireless sensor network platforms developed by Intel Corporation. Imote uses 32-bit ARM7 processor, 64KB SDRAM, 512KB Flash. Imote2 is an upgraded version of Imote. Imoete2 integrates PXA271Xscale CPU and RF chip compatible with IEEE802.15.4 [2]. Imote2 integrated Intel PXA271 CPU can work in low power mode, low voltage 0.85V, low frequency 13MHz. Imote2 uses dynamic voltage scaling technology, the frequency range can be from 13MHz to 416MHz, and supports a variety of different low-power modes, such as sleep and deep sleep modes. PXA271CPU integrates 3 memory chips: 256KB SRAM, 32MB SDRAM and 32MB Flash. Imote2 provides a variety of I/Os that can flexibly support different kinds of sensors, A/D conversion modules, and RF modules. The I/O of Imote2 includes: I2C, 2 synchronous serial ports, 3 high-speed UARTs, GPIOs, USB Client, USB Host, I2S audio coding interface, infrared interface, PWN pulse width modulation, camera interface, high-speed bus (Mobile Scaleable Link )interface. The PXA271 includes multiple timers and clocks. In addition, Imote2 adds 30 new DSP media processing instructions, supports video operations, and is compatible with Intel MMX and SSE integer instructions. In terms of wireless communication, Imote2 uses TI's CC2420 IEEE802.15.4 transmitter, which supports 2. 4GHz bandwidth 16 channels 250Kb/s data transmission rate, standard receiving range is 30 meters. The power supply part of Imote2 is powered by 3 AAA batteries, and rechargeable batteries can also be used, with a maximum current of 500mA. In terms of software, Imote2 supports TinyOS and Linux systems (currently supports TinyOS, and its data sheet has introduced that it will support Linux systems in the future). Typical application areas: police emergency on-site monitoring, industrial and agricultural monitoring, civilian community and parking lot monitoring, field video monitoring. Imote2 is currently offered as a commercial kit by MEMSIC Corporation.
c. CMUcam3 The
CMUcam3 node is a video sensor node introduced in 2007 by the team of Anthony Rowe of Carnegie Mellon University. The hardware of CMUcam3 node consists of CMOS image sensor OV7620, memory AL4V8M440 and microprocessor LPC2106, consisting of three modules in total. The CMUcam3 node can perform various digital image processing on the collected images, such as image difference, image convolution, and image compression. The CMUcam3 node supports two low-power modes: Idle and Power Down. The storage capacity of the CMUcam3 node is 64KB RAM and 128KB ROM. CMUcam3 is open source and there is no commercial kit available on the market.
d. DSPcam
DSPcam node [11] is a high-performance video sensor node fully upgraded on the basis of CMUcam3 node. The DSPcam node adopts a modular design, and each module has a separate board, which is convenient for function expansion and upgrade. The core microprocessor of the DSPcam node is a 32-bit Blackfin series DSP processor. The camera uses 1.3V ultra-low voltage power supply OV9653 and has an IEEE 802.11b/g wireless communication module. The DSPcam node supports the DMA function. DSPcam node embeds µCLinux real-time operating system (Real Time Operating System, RTOS), and customizes the MAC layer protocol based on priority assignment and provides QoS guarantee, which can provide real-time guarantee for the transmission of real-time video streams. Also DSPcam has no commercial kit.
e. XYZ
The XYZ node [9] is a video sensor node introduced in 2007 by the Lymberopouls team at Yale University. XYZ node adopts 32-bit ML67Q500X processor, CMOS image sensor OV6720, ZIGBEE wireless transceiver CC2420 and integrated temperature, humidity and other scalar sensors. In terms of software support, the XYZ node supports the SOS operating system and the IEEE802.15.4 protocol. The XYZ node is open source and there are currently no commercial packages on the market [10].
f. MeshEye
Stephan Hengstler's team at Stanford University proposed MeshEye, an intelligent node design scheme for video surveillance, in 2007. MeshEye adopts 32-bit Atmel AT91SAM7S processor, 64KB RAM and 256KB Flash, and adopts TI's CC2420 chip Zigbee transmission protocol. The MeshEye node uses two cameras with different pixels. Two cameras with different pixels are used together, the 30x30 pixel low pixel camera ADNS-3060 will wake up the high pixel VGA camera ADCM-2700 to capture the scene only when it detects a moving scene. In terms of software support, MeshEye does not support the operating system, and the program runs directly on the bare metal. There is currently no commercial kit.
g. WSN430/M3/A8
WSN430/M3/A8 node platforms are all open source WSN hardware platforms developed and provided by IoT-LAB [13]. IoT-LAB is an ultra-large-scale open test platform suitable for small wireless sensor network device testing, as well as heterogeneous wireless sensor network communication testing. IoT-LAB uses two hardware node platforms, the first is the WSN430 node for perception and the second is the M3 and A8 nodes for gateway and control functions. These nodes are all open source, and relevant information can be obtained on GitHub [14].
The WSN430 node is built based on the low-power processor TI MSP430F1611, adopts the IEEE 802.15.4 standard communication protocol, and the channel frequency is 2.4GHz. MSP430F1611 is a 16-bit CPU, 48KB Flash, 10KB RAM, external memory chip STM25P80, capacity is 1MB, power supply 3-7v, 830mAh. The WSN430 platform integrates a visible light sensor as well as a temperature sensor. The software supports TinyOS, Contiki and FreeRTOS operating systems.
The M3 node is built based on the STM32F103REY microcontroller (ARM Cortex-M3 core), and the communication adopts the IEEE802.15.4 standard, 2.4GHz. The STM32F103REY microcontroller is a 32-bit CPU, 72MHz main frequency, 64KB RAM, and external 128MB Flash. Power supply 3-7v, 650mAh. Software supports Contiki, FreeRTOS and Riot operating systems.
The A8 node is the most powerful node platform of this IoT-LAB, supporting complex operating systems such as Linux. The A8 node uses two 32 processors, the main processor TI SITARA AM3505 (ARM Cortex-A8 core), and the slave processor STM32F103REY. The A8 node platform can run complex programs, similar to those running on set-top boxes, smart phones, and tablets, to collect sensory information from wireless sensor networks. In terms of wireless communication, A8 adopts the IEEE802.15.4 standard, the channel is 2.4GHz, and provides USB and Ethernet interfaces. The A8 platform integrates GPS devices, accelerometers, magnetometers and gyroscopes. The software supports Linux system. These three nodes have powerful processors and storage performance, but the disadvantage is that the communication bandwidth is limited and the ability to transmit video images is limited. At the same time, these three node platforms have no kits available on the market.
Compared with low-performance platforms, high-performance WSN platforms have greatly improved processing capabilities, storage capabilities, and transmission capabilities. The high-performance platform can collect, process and transmit image and video information, and is suitable for more fine-grained and accurate information monitoring applications. There are many high-performance platforms, such as Cyclops[15], Panoptes[16], Meerkats[17], FireFlyMosaic[18], MicrelEye[19], CITRIC[20], etc. The literature [21] has more details on these platforms description and introduction. The high-performance platforms described above, with the exception of iSense and Imote2, which currently have commercial kits on the market, are designed by researchers for specific functions and applications. The main purpose of these platforms is to verify or achieve a certain goal of the researcher rather than commercial application. Therefore, for application developers, many detailed parameters of using these platforms are not known, and there are no kits available on the market.
Comparison & Summary: High-performance platforms mainly focus on several parameters such as processor type, storage capacity, interface with peripherals, and operating system.
4 Application status and analysis:
Monitoring and transmitting physical scalar information, such as temperature, humidity, illumination, acceleration and other information, requires low computational complexity and occupies a relatively small storage space, and a low-performance hardware platform can meet the requirements.
If it is necessary to transmit information such as images and videos with a large amount of data and high computational complexity, a high-performance platform must be selected.
Memsic (formerly Crossbow) dominates the market for low-performance hardware platforms. The MICA series node platform provided by Memsic is widely used in the development of WSN. Many academic research as well as industrial agricultural applications use Memsic products. Memsic's products are well coupled with WSN's software. For example, the TelosB node was developed by the University of California, Berkeley, and licensed to Memsic for commercial production. Its supporting operating system, TinyOS, is also developed by the University of California, Berkeley. In the past ten years, TinyOS has become the most widely used software operating system in the WSN field. Due to the matching of software and hardware, and TinyOS is open source and has strong development community support, TelosB platform is widely used in WSN research and application systems. According to the survey in the past three years, several important international academic conferences on WSN, such as MobiHoc, IPSN, Sensys and EWSN, most of them research WSN related technologies, such as routing protocol, MAC protocol, positioning technology, etc., the hardware platform used Most of them are Memsic's MICA series and Telos series, of which TelosB is the most frequently used. It can be seen that the low-performance platform represented by TelosB has a large group of loyal users in academia.
In terms of high-performance hardware platform applications, there is currently no large-scale application in the market. The application of the high-performance hardware platform is mainly an expansion of the information collected by the low-performance hardware platform. The most typical one is to provide visual video image information, which makes the monitoring information more diversified and can provide more fine-grained and accurate monitoring applications. But at the same time, there are many factors that restrict the large-scale application of high-performance hardware platforms, such as high cost, high power consumption, high transmission bandwidth requirements, and technical problems in energy supply, network transmission, and system integration. It is not completely solved, which limits its application promotion. Therefore, the existing high-performance hardware platforms are basically limited to the stage of laboratory use. The high-performance hardware platforms introduced above, except for iSense and Imote2, which have commercial kits, are all developed for specific research purposes and have not formed commercial production. In addition, a phenomenon is that in recent years WSN-related conferences, research papers on the use of high-performance hardware platforms are relatively rare.
Research on WSN, including the use of low-performance hardware platforms and high-performance hardware platforms, has a history of more than ten years. The research of WSN has derived many specific categories, such as wireless body sensor network (WBSN or BSN) [22], wireless multimedia sensor network (WMSN) and so on. For the hardware platform of WSN, according to the existing data, it was developed nearly ten years ago, and few new hardware platforms have been commercialized and applied in the past two or three years. For example, Crossbow, a well-known provider of WSN hardware platform, has not launched any new hardware platform related to WSN so far after being merged by Memsic. In addition, WSN research in the strict sense has not seen a major breakthrough. The following figure shows a development history of wireless sensor network hardware platforms. It can be seen from the figure that the WSN hardware platform has been designed since the 1990s. In the first ten years of this century, new products were born every year. However, in the past two or three years, no more representative ones have been seen. related new products.
The development board chips found in recent years:
cc2530: 8KB RAM , 256KB Flash,
cc2538 : MCU based on ARM Cortex-M3, 32KB RAM, 512KB flash memory , IEEE802.15.4 radio frequency function, can run 6lowpan protocol, coap protocol, mqtt protocol, and support Thread, ZigBee, released in 2013
cc2630 : Based on 32-bit ARM Cortex-M3, 2.4-GHz, 128KB flash memory, 28KB RAM, standby current 2mA, sleep current 0.1uA, that is, the average power consumption of waking up every two seconds is 17.4uA, 2 Powered by dry batteries, up to 10 years
cc2650 : Based on 32-bit ARM Cortex-M3, 2.4-GHz, 128KB Flash, 28KB RAM, 6lowpan, Bluetooth v4.1, Zigbee
cc2652R : Based on ARM Cortex-M4F, 2.4-GHz, 352KB flash, 80KB RAM, Thread, Bluetooth, Zigbee, chip released in 2015
2640 is mainly used for bluetooth application development
2652R is used for thread application development
The CC2538 works with TI's free Z-Stack software solution to provide the most powerful and stable ZigBee solution on the market.
The core of the cc2530 is the 8051 single-chip microcomputer, which is the architecture of the 1970s . There are many unstable factors in the application of smart home . The cc2630 makes up for all the defects of the cc2530. With routing function, cc2650 can not directly act as routing and coordinator
Link: https://blog.csdn.net/mzy202/article/details/53462614
Summarize:
2530, 2538, 2630, 2650 are relatively popular development boards that support zigbee and 6lowpan. 2538 is launched for smart home and smart energy. It has high-efficiency processing, low power consumption, large programmable space, and is relatively stable. more applications
Therefore, it is recommended to use a development board kit based on the cc2538 chip
There are more development boards sold on Taobao: cc2530 kit, cc2538 kit.
CC2530 Gateway Development Board
Taobao:
cc2538 Gateway Development Board ---6LoWPAN IoT gateway function (such as border router) for local 6LoWPAN network, thus linking wireless IPv6 network to servers on the Internet. The NAT64 translator translates between the 6LoWPAN IPv6 sensor network protocol and IPv4 used on the Internet.
Development kit based on cc2538 ---abundant sensor types
Further development based on 6lowpan: thread protocol
Thread protocol: Home IoT communication protocol technology Thread
Google's Nest Labs, proposed in July 2014, is an IP-based wireless network protocol used to connect smart products in the home.
Background:
The more commonly used network protocols are WiFi, Bluetooth, ZigBee, Z-Wave, but they all have shortcomings:
WiFi consumes a lot of power and is suitable for transmitting a large amount of data
Bluetooth power consumption is relatively low, but there is still a chaotic situation where Bluetooth 2.0 and Bluetooth 4.0 coexist, and IPv6 is not supported
ZigBee technology is more complicated and the R&D cost is high
Z-Wave is dominated by the Danish company Zensys and is not as strong as the ZigBee alliance for the time being
