1.Android camp 3D sensing internal and external difficulties, MediaTek, and joining the game;
According to the micro-network news, last year Apple's iPhone X was equipped with Face ID function, which raised the industry's great concern for 3D sensing. This year, Android camp vendors such as Huawei and Xiaomi are expected to follow up. Tuoba Research Institute expects that by 2020 The output value of the global smart phone 3D sensing module will reach 10.85 billion U.S. dollars.
Previously, foreign media reported that the supply of 3D sensing key VCSEL devices is still insufficient. The fastest time to 2019 is when 3D sensing will be available on Android smartphones. Pierre MEMS and imaging technology research director at market research institute Yole Developpement Pierre Cambou said that because Apple’s TrueDepth camera technology sets a high threshold, he predicts that other competitors may take more than a year to provide iPhone 3 with comparable 3D sensing technology. Therefore, at this stage, the Android camp mainly chooses an under-the-screen fingerprint identification program. Such as the released Vivo X20 Plus UD, as well as the release of the millet 7, Meizu 16, Samsung Note 9, and so on.
In spite of this, it is expected that the Android camp will still follow up and carry the 3D sensing module.
The most critical and challenging part of 3D sensing technology is the ability to accurately measure distances, and different algorithms have formed patent barriers, which also cause slight differences in the components of each 3D sensing module. Sensing technologies include Stereo Vision, Structured Light and Time of Flight (ToF).
In addition to the Lumentum/II-VI of the Apple camp, there are not many suppliers with complete solutions (capable of providing modules) in the 3D sensing technology supplier camp. Currently 3D is shipped. The test module is to the mobile phone manufacturer's Google camp. The German PMD designs IR CIS and Infineon OEM. The key VCSEL is provided by Princeton Optronics. The other camp is Qualcomm+Himax (Qijing Optoelectronics) solution provided by Qualcomm and Chip design, Wonderscape provides WLO and DOE optical devices.
The Qualcomm+Himax solution is currently a mature Android 3D sensing solution. Himax masters the solution core algorithm and hardware design and manufacturing capabilities, and provides a fully integrated 3D sensing solution for structured light modules (SLiM, Structured Light Module). Electronic report, Himax participated in the design of most of the components, self-made, including homemade DOE and WLO, design ASIC, CIS, laser emitter IC, and the integration of the entire Tx module, in which ASIC embedded high-pass 3D depth map generation Algorithm. At the same time, Himax also independently designed the AA equipment for Tx end assembly. It is expected that SLiM (structured light) production capacity will reach 2kk/month at the end of 18Q1, which is mainly digested by the mainland mobile phone brands. In addition to SLiM, Himax currently provides 3D for its customers. Sensing components/solutions include WLO (for Apple), Qualcomm+Himax SLiM (for Android high-end machine), binocular vision solution (for Android low-end machine), WLO has mass production. Binocular vision program is mainly applied two cameras Simulate 3D vision and use coded light to enhance image depth information. Himax's binocular vision 3D sensing solution is targeted at low to mid-range Android machines, priced at less than $10.
Himax is a leading manufacturer of display driver ICs in Taiwan. Since its launch in the US in 2006, it has continued to expand new fields of research and development, including CMOS image sensors, LCOS microdisplays, etc., and subsequently joined hands with Qualcomm to enter 3D sensing, including Rx-side CIS dimensions. It is only 20% of the ordinary mobile CIS modules, more than 33,000 projection spots, and the error rate of less than 1% in the range of 20 to 100cm. It can be said that it is the highest quality 3D sensing SLiM solution in the Android camp. In addition, Himax started supplying WLO products to Apple in the second half of last year. The subsequent outbreak of SLiM and the rapid development of the entire industry will also drive the company's WLO products to increase rapidly.
Haitong Electronic Research reported that Himax has the strength to provide a complete solution and participate in most of the core device design or manufacturing, mainly based on its deep accumulation in the field of NIR CMOS sensors, precision manufacturing capabilities in the field of devices such as WLO/DOE. As well as the ability to assemble and test laser emitter modules, Himax has a deep knowledge in the field of 3D sensing. Whether it is program maturity, production schedule, or core algorithms, component design, and program design, Himax is at the forefront of the Android camp. It is expected to fully enjoy the feast of the industry.
VCSEL's major suppliers are controlled by Apple, and Android camp is the best for them
Comparing the supply chain of iPhone and Qualcomm+Himax solutions, we can see that the key components of 3D sensing are VCSELs, which are commonly used by ToF and Structured Light, and Edge-Emitting Laser (EEL) is another One solution, however, is that VCSELs have better beam quality, lower beam divergence, lower energy consumption, and more module size. iPhone uses VCSEL, and Qualcomm+Himax uses EEL.
VCSELs are manufactured using semiconductor manufacturing processes. The Apple camp uses 6-inch gallium arsenide wafers for cutting. However, there are not many manufacturers currently achieving 6-inch mass production in the industry; most other VCSEL suppliers currently produce 4 inches of energy, and a few are With 3 inches, the market's overall supply and demand situation is tight, and it also affects the speed with which non-Hilton camps can import 3D sensing technology.
In addition, there is a patent agreement between Lumentum, the main supplier of VCSELs, and Apple, which allows Android camps to follow up on VCSELs and choose EEL in the short term. However, EEL's photoelectric conversion efficiency is poor and the cost is high, which will make Android The camp's 3D sensing solution is still difficult to compare with Apple in terms of efficiency and cost. Compared with ordinary front camera, carrying 3D sensing module will increase the cost by 20 US dollars to 25 US dollars, Huawei, OPPO, vivo, millet and other mobile phone manufacturers are It is planned not to pursue a large quantity of high-end models. It is only possible to use the mainstream model after the cost has dropped significantly. It is expected that it will wait until 2019.
According to TPI, it is conservatively estimated that in 2018, only two Android vendors may follow up, including Huawei and millet, but the number of production will not be too much, so Apple will still be 3D sensing for mobile phones. The largest adopter of mobile phones is expected to have 197 million smartphones equipped with 3D sensing modules worldwide in 2018, of which iPhones will account for 165 million. In addition, 2018's 3D sensing module market output is estimated to be approximately 5.12 billion U.S. dollars, of which the proportion contributed by the iPhone is as high as 84.5%.
Three major challenges: production capacity, yield, and cost. MediaTek is ready to join.
In the face of rapidly growing market demand, on the one hand, the capacity of the Qualcomm + Himax solution cannot be far from satisfied. On the other hand, the debugging of the 3D sensing module is also a problem. According to the industry chain at the beginning of the year, there are a few 3D cameras for the module plant. After the module has sent samples to Qualcomm for one or two months, it still has not received the signal that the commissioning has been completed. The module manufacturers have to wait while waiting for them, and constantly improve the yield rate. Since the debugging progress of Qualcomm is not ideal, the mass production progress may be Some delay.
At present, the main chip of the Android camp 3D sensing module is basically provided by Qualcomm. In the face of huge market potential, MediaTek is already ready to enter. It is reported that MediaTek also intends to join the 3D sensing battlefield with the role of APU supplier, intending to use a convolutional neural network. (CNN) - similar to Apple's neural engine - to support biometric identification. Insiders said that MediaTek will be combined with the 3D camera designed by Opie Zhongguang to provide CNN accelerator for Xiaomi in the future.
At this year's MWC, MediaTek displayed a mobile 3D sensor camera, and the new P-series chip platform will support Aobi Zhongguang 3D sensor camera. In 2016, MediaTek took over Ophir Zhongguang, its platform reference design and The latter's 3D sensor technology is fully adapted. After two months of the iPhone X release, Aobi Zhongguang sent the module to the mobile phone manufacturers of MediaTek and domestic TOP 3, thus, Opie Zhongguang. Also became China's first manufacturer to send sample phone front 3D camera.
In addition, another major IC design company in Taiwan has also dug several R&D staff from Qijing Optical. They are all core team members of the original Google Glass development project.
Qualcomm+Himax, MediaTek+Obizhong Zhongguang, and the combination of the latter's introductory players will enable the 3D sensor technology of the Android camp to go even further. Will the industry be as good as the iphone X? The industry will wait and see.
2. Pass three new iPhone face recognition standards;
Set micro-network news, Apple announced three new iPhones this year almost become a foregone conclusion, the most concerned is the first time in the iPhone 3 appear 3D sensing face recognition capabilities, is expected to be imported into the standard equipment in all three new aircraft this year.
Most of the past 3D sensing uses IR LEDs for detection, but the accuracy is still difficult to meet smart phone applications. Apple uses a more accurate VCSEL (Vertical Resonant Cavity Surface-Laser) technology, whose sensing components are dominated by Lumenum, Finisar. Development, new to provide epitaxial, stable, Hong Jieke is responsible for OEM, and then build a 3D sensing ecosystem.
Recently, the industry has circulated multiple spy photos of three new iPhones. The screens all have “bangs”, which means that the face recognition system may become the standard.
The market has already rumored that the new iPhone will be built with a full-screen design this year, and it will have a “bangs” shape and Face ID function that will help 3D sensing applications become more popular. This will also help the GaAs market demand. Double growth.
After the iPhone X was listed at the end of last year, the face recognition application has become one of the most popular new applications for smartphones, and the key component 3D sensing components that realize this application have been regarded by the market as the most embarrassing industry this year.
3. Stabilize edge node communication as the primary task IIoT realizes network edge intelligence;
Industrially-networked machines can sense a wide variety of information that can be used to make critical decisions in the Industrial Internet of Things (IIoT) environment. Sensors located in edge nodes can be far from any Data Aggregation Point in the process. In the middle, it must be linked through a gateway, and this kind of gateway is mainly responsible for transmitting edge data to the network.
The sensors form the front end of the IIoT system. We measure the data and convert the sensed information into quantifiable data, such as pressure, displacement, or number of rotations. After the data is filtered, the most valuable information is selected. Then it is sent back from the node to the back-end system for processing. The low-latency connection allows the system to make critical decisions immediately after receiving critical data.
Edge nodes generally must be connected to the network through wired or wireless sensor nodes (WSNs). In this signal chain, data integrity is still very important. If the communication is not continuous, the wire breaks or the quality degrades, optimizing sensing and measuring. The data has no value at all. When designing the system architecture, the first consideration is the robust communication protocol. The best choice depends on the link requirements, including distance, bandwidth, power, interoperability, security, and reliability. Sex.
Wired industrial communication plays a key role in technologies that require on-line stability such as EtherNet/IP, KNX, DALI, PROFINET, and ModbusTCP. Setting range Sensor nodes in every corner of the plant are using wireless networks and gateways. Communication, and the gateway relies on the wired infrastructure to link to the main system.
Sensor nodes must have networking capabilities
Only a handful of networked IoT nodes will use wired communication alone in the future. Most of these devices will use wireless networks. The highly efficient industrial IoT linking strategy must allow sensors to be located wherever valuable information can be sensed. In the area where the communication and power devices are currently installed.
The sensor node must have a method of communicating with the network. With the higher-order communication protocols mapped to this type of link by the industrial Internet of Things framework, it is expected that the wired communication part will still follow the Ethernet network. The Ethernet network is set up from 10Mbps. Covers transmission rates above 100 Gbps. The higher-order speed parts are usually targeted at the backbone line between the Internet connection to the cloud server host cluster.
Industrial networks such as KNX, which are slower, use twisted-pair copper to transmit differential signals, use 30 volts of electricity, and have an overall bandwidth of 9600 bps. Since each segment can support a limited number of addresses (256), Therefore, the addressing mechanism can support up to 65,536 devices. The maximum transmission distance of each network segment is 1,000 meters. The user can choose to configure the repeater. Each repeater supports up to four network segments.
Industrial Environment Wireless Networks Face Multiple Challenges
When the IIoT wireless network system designers are considering which communication and network technologies to adopt, they will find themselves facing many challenges. They must consider the following restrictions from a higher position:
. Transmission distance
Intermittent or continuous link
Bandwidth
Power
Interoperability
. safety
Reliability
Transmission distance
The so-called distance here refers to the distance traveled by the data transmitted by the connected IIoT device. The short-distance personal area network (PAN) has a transmission distance of a meter class (Fig. 1), such as a Bluetooth Low Energy (BLE) technology A local area network (LAN) with a transmission distance of several hundred meters can be used to install various automated sensors in the same building. As for a wide area network (WAN) with a transmission distance of several kilometers, Its application includes the installation of various agricultural sensors in a vast farm.
Figure 1 Short-range wireless link
The selected network protocol should match the transmission distance required by the industrial IoT context. For example, for an indoor LAN application with a distance of several tens of meters, the 4G mobile network is not suitable in terms of complexity and power. When data transfer distances are challenged, edge operations are a viable alternative. We can perform data analysis directly at the edge nodes without having to send data back to the main system for processing.
The power intensity of transmission radio waves is inversely proportional to the square of the transmission distance. The signal power intensity is inversely proportional to the square of the distance traveled by the radio waves, so when the transmission distance is doubled, the radio power received by the receiving end is only one fourth of the original power. For every 6dBm increase in output signal power, the transmission distance will double.
In the ideal barrier-free transmission space, the inverse square law is the only factor that affects the transmission distance. However, the transmission distance in the real world will be attenuated due to the obstruction of objects such as walls, fences, and plants on the transmission route.
In addition, moisture in the air also absorbs RF energy. Metal objects reflect radio waves, causing secondary signals to arrive at the receiver at different points in time, and additional power loss can also cause destructive interference.
The radio receiver sensitivity determines the maximum propagation path loss. For example, in the 2.4 GHz Industrial/Scientific/Medical (ISM) band, the minimum receiver sensitivity is –85 dBm. The RF radiation propagates uniformly in all directions, the intensity contours. It will form a sphere (A = 4πR2), where R is the distance from the sender to the receiver, in meters. According to the Friis transmission formula, free space loss (FSPL) and sender and receiver The square of the distance between them and the square of the radio signal frequency are proportional.
In the formula, Pt = transmission power in watts, S = power at distance R.
In the formula, Pr = received power in watts.
λ (wavelength of transmission signal, in meters) = c (light speed) / frequency f (Hz) = 3 × 108 (m/s2)/f (Hz) or 300/f (MHz)
Where f = transmission frequency
If the transmission frequency and the distance to be transmitted are known, the transmission and reception data can be calculated according to the FPSL. The link budget is shown in Equation 1.
Received power(dBm)=Transmitted power(dBm)+gains(dB)–losses...... Formula 1
Bandwidth and Links
Bandwidth refers to the rate at which data is transmitted within a unit of time. Bandwidth limits the maximum rate at which IIoT sensor nodes collect data and transmit data. The factors considered are as follows:
The total amount of data generated by each device after a certain period of time.
The number of nodes deployed and aggregated to a gateway.
Considering the continuous or intermittent peak transmission mode, how much available bandwidth is needed to meet the peak demand.
The packet size of the network protocol should be matched with the amount of data to be transmitted. If the packet is filled with blank data, the efficiency of this protocol is not high. However, the larger packets are finely divided into many smaller data packets and transmitted separately. Paying for the cost of resources. The IIoT device does not connect to the network at any time. It will only go offline after the data has been transmitted at regular intervals to save power or bandwidth resources.
Power and interoperability
If the IIoT device using battery needs to save power, then the device must immediately switch to sleep mode as soon as it is idle. We can start to adjust the power consumption mode of the device according to different network load conditions, which will help the power supply of the device. The battery capacity can be matched with the power required to transmit the necessary data.
The possibility of interoperability between various nodes in the network is bound to become a major problem. The industry's traditional approach is to use standard wired and wireless communication protocols in order to maintain interoperability within the Internet. Emerging IIoT products due to the need to cooperate with the new Release the rapid pace of technology, and the work that leads to standardization is fraught with difficulties. The IIoT industry system is based on the best technologies and these technologies are related to commercially available solutions. If the technology is widely adopted by all walks of life. , then the probability of achieving long-term interoperability will be higher.
Safety
IIoT network security plays an important role in the system in three aspects: confidentiality, integrity, and authenticity. To maintain confidentiality, network data must be in a well-known framework and cannot be leaked to external devices or externally. Device interception.
In order to maintain data integrity, the content of the signal must remain exactly the same as it was at the time of issue, and cannot be altered, truncated or added. As for maintaining authenticity, receiving data must be determined from the intended source, excluding other sources. Messages and false nodes for false communication are examples of loss of authenticity.
Even a secure wireless node, once interfacing to a non-secure gateway, can create loopholes that allow interested people to get an intruder's breach. The data timestamp can help identify whether the signal has gone through frequency hopping. And retransmission through the side channel (Side Channel). Time stamps can also be used to correctly reorganize out-of-order key data, allowing packets to go through numerous non-coordinated sensors and then restore the original data.
Security support for the AES-128 encryption standard, which follows the IEEE 802.15.4 and AES-128/256 specifications within IEEE 802.11. Key management, garbled generation (RNG), and Network Access Control List (ACL) These all help to increase the security barrier of the communications network.
Band
Some IoT wireless sensors will use the radio frequency band in the mobile phone infrastructure. However, such sensors are usually devices with high power consumption. One example is the vehicle-carriage communication system. If such a system wants to collect information on actions It is practically not feasible to transmit over short-range wireless communication technology. On the other hand, many other low-power industrial applications use unlicensed bands in the ISM band.
The IEEE 802.15.4 low-power wireless communication standard is an ideal technology for many mission-critical IoT applications. The frequency bands used include the 2.4 GHz, 915 MHz, and 868 MHz portions of the ISM band. A total of 27 channels are available for multi-frequency channel frequency hopping. Use (Table 1).
Unlicensed bands available around the world have inconsistent physical layer support. Europe provides 600 kHz wide Channel 0 channels at 868 MHz, while North America offers 10 2 MHz wide bands at 915 MHz. The rest of the world Provides 5MHz wide Channel 11 to Channel 26 at 2.4 GHz.
Low-power Bluetooth provides a solution with greatly reduced power consumption. Bluetooth low energy is not suitable for transferring files and is suitable for transmitting small amounts of data. One of the advantages of Bluetooth low energy is that the penetration rate is much higher than that of other competitors. Has been widely integrated into a variety of mobile devices. Bluetooth 4.2 core specifications using the 2.4GHz ISM common frequency band, transmission distance of 50 to 150 meters, the use of Gaussian (Gaussian) frequency shift modulation mechanism can achieve 1Mbps data rate.
When deciding which optimal frequency to use for an IIoT solution, the advantages and disadvantages of the 2.4 GHz ISM solution should be taken into account:
advantage
In most countries, it is not necessary to obtain a license.
The same solution can be sold in various markets.
The bandwidth of 83.5MHz is enough to be divided into multiple channels, and transmitted through multiple channels to achieve high data transmission rate.
Duty Cycle can reach 100%.
The antenna size is smaller than the 1GHz band antenna.
Shortcomings
At the same output power, the transmission distance is shorter than the 1GHz band.
High permeability generates many interference signals.
Protocol
In the communication system, a set of rules and standards will be used to regulate how data is formed and how to control the exchange of data. For example, the Open Systems Interconnection (OSI) model divides communication into multiple functional layers to make it easier for people to build and expand. The interworking network. OSI model is divided into 7 layers (Figure 2), including entity (PHY), data link, network, transmission, conversation, expression, and application layering.
Figure 2 OSI and TCP/IP Model
The IEEE 802.15.4 and 802.11 (Wi-Fi) standards specify the media access control (MAC) data link sublayer and physical layer. 802.11 base stations that are close to each other may use one of the non-overlapping channels, respectively, to reduce interference effects ( Figure 3). The modulation mechanism used by 802.11g is orthogonal carrier frequency division multiplexing (OFDM). In the following we will introduce a more complex mechanism than IEEE 802.15.4.
Figure 3 The global general IEEE 802.15.4 physical layer Channel 11 to Channel 26 and the IEEE 802.11g Channel 1 to Channel 14 channels
The link layer provides a mechanism for converting radio signals into bit data and converting bit data into analog signals. This layer is also responsible for performing reliable communications and managing radio channel access operations. The network layer is responsible for controlling data on the network. Paths and Addressing Jobs Passed In this layer, the Internet Protocol (IP) is responsible for providing IP addresses, as well as passing IP packets from one node to another.
When running an application session on both ends of the network, the transport layer will generate a corresponding communication chat program. This design allows a device to run multiple applications at the same time, and each application uses its own communication channel. The connection on the Internet Most network devices use Transmission Control Protocol (TCP) as the default transmission protocol.
The application layer is responsible for the format and control of the data, allowing the node sensor's specific application to optimize its transport data flow. One of the most popular application layer protocols in the TCP/IP stack is the Hypertext Transfer Protocol (HTTP). This protocol was developed for passing data over the Internet.
The FCC Part 15 rule of the United States Federal Communications Commission limits the effective transmission power of the ISM band to 36dBm. One of the exceptions is to use a 24dBi gain antenna and a 24dBm transmission power for a fixed point-to-point link using the 2.4GHz band. Isotropic RF power (EIRP) up to 48 dBm. Transmission power should be at least 1 mW. To make the packet error rate less than 1%, the receiver sensitivity should be able to receive the –85 dBm signal in the 2.4 GHz band and receive 868 MHz. Signal strength of –9dBm on the 915MHz band.
Old building or new ground settings
Industrial IoT must be supported by a large number of wired and wireless standards before it can be put into operation. However, there are not many current options for constructing IIoT using existing network systems. Newly developed IIoT solutions must be adjusted to integrate into the network environment.
Greenfield is to create a new system from scratch in a brand new environment. There will be no restrictions and restrictions imposed by old equipment. For example, to build a new factory or warehouse, consider installing IIoT solutions in the steel structure of the building. To achieve the best performance.
Older facilities (Brownfield) installed the IIoT network in the existing infrastructure and the challenges will be more severe. The old network may not be suitable for running the IoT, but the new IIoT system must be installed with any Systems co-exist, and these old systems are often the source of RF interference signals. Developers must inherit the constraints imposed by hardware, embedded software, and previous design decisions in the old environment. The development process therefore becomes extremely tedious. Carefully and carefully analyze, design and test.
Network topology
The IEEE 802.15.4 protocol provides two device classes. The full-function device (FFD) can be used in any topology and can communicate with any other device as a PAN coordinator. The RFD can only be installed in the star. Topology, and cannot be used as a network coordinator. Only one network coordinator is required in the simple build environment of the IEEE 802.15.4 specification. The user can select a suitable network model based on the application form, including peer-to-peer. ), Star, Mesh, and Multihop (Figure 4).
Figure 4 Peer to peer, star, mesh and multi-hop topology
The peer-to-peer topology network simply links two nodes, but does not use any intelligence to expand the network link distance. This type of topology is fast to set up, but once there is a node failure, the entire network will be shut down, and there is no redundancy at all. Words.
The star topology extends the distance of the radial network and increases the length of the transmission between the two nodes. As with an FFD node, the master can communicate with multiple RFD nodes, but each RFD node is still only Can communicate with the router. As long as it is not FFD, even if there is a Single Point of Failure in this topology, the entire network can continue to operate.
The mesh network topology allows any node to skip other nodes to communicate with each other, thereby providing redundant communication paths to increase the strength of the network. The intelligent mesh topology network can communicate with the least jumping path to reduce power consumption and transmission delay. This topology with a Self-Organization mechanism can adapt to changes in the environment, allowing nodes to join or withdraw from the network freely.
reliability
IIoT users are most concerned with reliability and security. Organizations often rely on large and complex clusters to perform data analysis, but these systems often have bottlenecks in data transmission, indexing, data capture, transformation, and load handling. To avoid bottlenecks in the downstream clusters, it becomes very important for each edge node to communicate efficiently.
The industrial environment is a very harsh place for high-efficiency radio frequency wave transmission. Large-scale, irregularly shaped, high-density metal plant equipment, cement walls, compartments, and metal shelves will all produce multipath electromagnetic wave transmission.
The radio waves are emitted from the transmitting antenna in all directions. “Multipath” refers to the situation in which the waveform changes after the radio waves have propagated through the environment (Environmental Propagation). The incident waves seen by the receiver are divided into three categories. For reflection, diffraction, and scattering, the waves transmitted by multiple paths may change in amplitude and phase, causing the destination receiver to see signals that are subject to constructive or destructive interference.
CSMA-CA channel access
Carrier Sense Multiple Access and Collision Avoidance (CSMA/CA) is a data link layer communication protocol in which network nodes use a carrier detection mechanism. A node only transmits once when it detects that a transmission channel is idle. The entire packet data. The hidden nodes in the wireless network are not in the detection range of other nodes. Figure 5 shows an example where nodes far from the edge of the transmission can still see the base station "Y" but cannot see the other node. X or Z.
Figure 5 The hidden nodes X and Z cannot communicate directly.
Handshaking program uses RTS/CTS to construct a virtual carrier sensing mechanism. It only needs to send out short request messages to transmit and clear data. This process transfers WLAN data. 802.11 mainly relies on physical carrier sensing. IEEE802.15.4 uses the CSMA/CA mechanism. To overcome these hidden node problems, the industry uses mixed RTS/CTS handshakes and CSMA/CA. When circumstances permit, increasing the transmission power of hidden nodes can lengthen the observed distance.
To improve the bandwidth, various advanced modulation mechanisms have been developed to modulate the phase, amplitude, or frequency of signals. Quadrature Phase Shift Keying (QPSK) This modulation mechanism uses four phases to encode each symbol into Two bits of data.
Use modulation mechanism to effectively improve bandwidth
Orthogonal modulation uses a hybrid architecture (Figure 6) to reduce signal bandwidth requirements by phase shifting. Binary data is split into two consecutive bits with the ωc carrier, sinωct, and cosωct trigonometric functions. The quadrature phase is modulated.
Figure 6 Offset QPSK modulation architecture
IEEE 802.15.4 transceivers operating in the 2.4 GHz ISM band use a QPSK-derived physical layer called offset QPSK, O-QPSK, or interleaved QPSK. A single data bit (Tbit) is added to the bit transport stream. The offset time constant, which shifts the data by half the symbol period, so as to avoid simultaneous transmission of the signal waveforms at node X and node Y, preventing the waveform from overlapping and causing interference. The continuous phase difference will never exceed the positive value. Negative 90 degrees (Figure 7). One of the disadvantages of O-QPSK is that it does not allow differential coding, but it does eliminate the problem of Coherent Detection.
Figure 7 Phase Transition ±90° (left) and I/Q O-QPSK Option (right)
The modulation mechanism adopted by IEEE 802.15.4 reduces the symbol rate for transmitting and receiving data. The O-QPSK modulation mechanism transmits two coded bits at the same time, with a symbol rate of 1 to 4: Bit rate. Therefore, 62.5 ksymbols/sec The symbol rate can reach 250kbps data transfer rate.
In response to the growth of the network, the addressing mechanism was expanded.
Not all IoT nodes require an external IP address. In terms of dedicated communications, sensor nodes should be able to support unique IP addresses. IPv4 supports 32-bit addressing mechanisms. This technology, developed decades ago, can only support 4.3 billion. The device, nowadays unable to respond to the needs of the Internet. IPv6 increases the addressing mechanism to 128 bits, and can support 240 times 10 times the 36th globally unique URL (GUA) device.
The need to map data and manage Web sites from two different IPv6 domains and IEEE 802.15.4 networks will pose a severe challenge to the design. 6LoWPAN defines encapsulation and header compression mechanisms to allow Ipv6 packets to pass IEEE 802.15. 4 network for transmission and reception.
One of the examples is Thread, which is a closed file but the license-free communication protocol can be run on 6LoWPAN basis to support various automation applications.
In response to this trend, semiconductor component suppliers such as Analog Devices Inc. (ADI) provide a full range of wireless analog transceivers supporting wired network protocols for AduCx series microcontrollers and Blackfin series DSPs. Such as low-power wireless high-power transceivers. The module scheme - ADRF7242, supports IEEE 802.15.4 protocol, provides self-configured data transmission rate and a variety of modulation mechanisms, and uses the globally-available ISM frequency band. Its transmission rate covers from 50kbps to 2000kbps, and can pass relevant US regulations FCC and European regulations ETSI standard certification.
Another product, the ADRF7023, uses the worldwide unlicensed ISM band, which includes 433MHz, 868MHz, and 915MHz. The transmission rate ranges from 1kbps to 300kbps. The company offers a complete WSN development platform that allows users to design custom solutions.
For example, the RapID Platform platform includes a series of modules and development kits that can be used to embed various industrial networking protocols. SmartMesh wireless sensors include multiple chips and pre-verified PCB board modules, and are equipped with mesh networking software for sensors Ability to communicate in a variety of harsh industrial Internet of Things environments.
(The author is ADI Automation Energy and Sensor Product Engineering Manager) New Electronics
4. Technical cooperation is the key interoperability of industrial IoT to improve IIoT deployment efficiency;
The Industrial Internet of Things (IIoT) is expected to revolutionize the way people operate manufacturing, energy, and transportation systems. However, due to the sheer size and complexity of the interconnection technologies that make up the Internet of Things, no company can independently provide a complete enterprise IIoT solution (Figure 1).
For a complete explanation, let's look at an IIoT system architecture. The IIoT system not only increases the number of smart devices and sensors, but also includes the transmission and management of large amounts of data through distributed networks (including edge nodes, local IT, and cloud). 2).
Figure 1 System designers need to integrate the technical components from various vendors' sources to build the systems needed for a particular application to improve quality, yield, efficiency, and safety.
Many different subsystems and technologies in the architecture of Figure 2 must be combined to build a complete solution.
To manage subsystems originating from various suppliers, communication among them is by no means a simple task. This is illustrated by the communication stack diagram of the Industrial Internet Consortium (IIC). We must manage not only Layer communication standards and protocols, and many vertical industries in the task industry (such as manufacturing or power grids), have their own set of industrial protocols to regulate (Figure 3).
Figure 3 There are still many traditional M2Ms deployed that use various proprietary protocols, which must also be integrated into the system.
Interoperability is the key to success
Because of this, when evaluating the technology of IIoT vendors, a key criterion is interoperability, which is the convenience of passing information back and forth across the technology boundary.
Information can be delivered in four ways: protocols, data files, Web services, and APIs. In IIoT systems, different parts of the system may use different methods. However, the ultimate goal is to make communication between subsystems as much as possible. Become simple, let the system designer can concentrate on solving real system problems, rather than solving the problems brought by the tool.
So, what standards should be considered when assessing interoperability? There are usually two aspects: Openness and technology partners.
Open Platform Improves Technical Function Limits
Openness refers to how easy it is for developers to use the platform to build and specify a system. When designing an IIoT system with multiple vendors, there are several features that allow users to program:
1. Supports many communication protocols, including various vertical industrial protocols such as CAN, Fieldbus, OPC UA, EtherCAT, Modbus and IEC-61850.
2. Supports multiple data file types.
3. Software Development Kit (SDK) and Module Development Kit (MDK), such as the PTC SDK for the ThingWorx Platform.
4. Open source real-time operating system, such as NI Linux Real-Time.
5. Open and extensible API.
6. Plug-ins and accessories, such as the LabVIEW Cloud Toolkit for Amazon Web Services.
These functions provide a variety of cross-technology data communication options, which can avoid system engineers from falling into design bottlenecks due to system development. Open platforms help to improve the limitations of technical functions or the limitations of only supporting one or two communication protocols.
Finding technology partners
In addition, partnerships between suppliers can provide integrated services, further reducing the risk of integrating adjacent technologies. Through joint efforts such as the IIC test platform, participating companies are integrating technologies from multiple fields and forecasting. Maintenance, typical IIoT application-building reference architectures such as smart grid communication and control.
These partnerships can be demonstrated through technology demonstrations at NI Industrial IoT Labs. The demonstration performs asset health monitoring on a pump and incorporates multiple vendor technologies, including:
1.Flowserve - Flow Control System Solution (Figure 4)
Figure 4 Flowserve Flow Control System Solution
2.Hewlett Packard Enterprise - Deep Edge Computing and Remote Management
3.NI (National Instruments) - Data Acquisition and Feature Extraction
4.PTC--IoT platform, including analysis and Augmented Reality (AR) capabilities for enterprise systems
5.OSIsoft - Data Management and History Library
Business leaders should be wary of suppliers who claim to provide complete IIoT solutions. Because a complete IIoT system will involve components from many fields of technology, from data acquisition to augmented reality. Conversely, the technology sought Partners should be aware of neighboring technologies and master the importance of effective integration and actively work with other vendors.
(This article is for NI Product Marketing Engineer) New Electronics
5. The researchers created the coldest electronic chip in the world
University of Basel professor Dominik Zumbühl and his colleagues succeeded in cooling the temperature of the nanoelectronic chip to 2.8 milliliters, which is approximately 273.15 degrees Celsius. The researchers said: 'Magnetic cooling is based on the principle that when the applied magnetic field gradually decreases Hours, the system will gradually cool, while avoiding any external heat flow.
'Before the magnetic field is reduced, the heat generated by the magnetization needs to be absorbed by other means in order to obtain effective magnetic cooling. This is how we successfully cooled the nanoelectronic chip to 2.8 milliKelvin to achieve a record low temperature. Method. '
Professor Zumbühl and his colleagues combine these two cooling systems, both of which are based on magnetic cooling.
They cooled all the conductive connections of the chip to 150 micro Kelvin (less than one thousandth of a degree from absolute zero).
They then applied the second cooling system directly to the chip itself and placed a Coulomb blockage thermometer at the same time. The structure and material of the thermometer allowed it to be cooled down to 2.8 milliKelvins by magnetic cooling.
Prof. Zumbühl said: 'We combined two cooling systems to cool the chip below 3 milliKelvin (about 273.15 degrees Celsius). We are optimistic that we can use the same method to achieve 1 milliKelvin.'
The scientists said: 'We were able to keep the chip at ultra-low temperatures for 7 hours, which is quite good. Scientists will have plenty of time to conduct multiple exploration experiments, which will help to understand the physics characteristics near absolute zero.' Science and technology
