IoT

A brief overview of IoT (Internet of Thing)

For years, IoT has been a hot topic. It is a computing concept that describes a future where every day physical objects will be connected to the Internet and be able to identify themselves to other devices. The term is closely identified with RFID as the method of communication, although it also may include other sensor technologies, wireless technologies or QR codes. IoT describes a world where just about anything can be connected and communicate in an intelligent fashion. In other words, with the Internet of Things, the physical world is becoming one big information system.
Internet of Everything (IoE) seems to be more frequently used by Qualcomm and Cisco. On the other hand, IoT is more likely a mainstream term. At this moment, both terms didn’t made much distinction between one another.

There are proven evidences that how IoT is going to shake the world economic, thus company must know how to grab the chance to move along with the emerging trend. Smart sensors, RFID, and cloud technology are exist for quite some time. This means that with the existing technology, Businesses can add value by combining the components to form a value-added products, by innovation and creativity, we can create infinite number of products to serve the market.

How it works? 

Let's take an example on door access control system, the workers use facial recognition to unlock the door. With integration between the access control and the HR system, we makes it possible to detach the HR department physically from the workplace, whether the HR staff can be worked from home or via shared-services provider, we can centralize control in a remote location. It has reduce the manual work on the supervisor. This is especially more efficient for company which has more than one factory location.

Staff's face scanned to unlock the door, this data was sent to the cloud with an assigned IP address (IPV6) , then HR department time check attendance record for payroll based on their based on these info. The management on the other end, base on these information to monitor. It benefits the HR, which they eliminate the work in collect each punch card in the end of the month, and also save workers from the hassle of carry the traditional RFID card with them every day. 

Motivation and key drivers

1. Implementing IoT is definitely for the interest of a better society. The rationale lies in the question of we are living in a dynamic world where things move so fast, how we cope with that. IoT will revolutionize the life we live, it change the way we do things in the day-to-day activities, and for example household appliances can be automated to deliver without human intervention. This concept turns into a renaissance, it increases the productivity of human being with the facility of IoT connected devices. We will be in the transition to a new society, where human can be the creator of unlimited creativity, we focus in the development of various IoT gadget and devices set to drive up the productivity. 
2. Higher employment rate as more human operator need to control the machines. The conventional point-of-view is Machine to machine (M2M) is going to take over human job, which turns out to be a higher unemployment rate. M2M is a broad label that can be used to describe any technology that enables networked devices to exchange information and perform actions without the manual assistance of humans. However, machine automation is lacking, it is still far from human consciousness to produce cognitive intelligence and emotions. Expert with optimistic point-of-view believe machines and computers cannot totally replace human insight, M2M is actually creating more opportunity for human work force. At this point of time, we can see there is still long way to go to fully integrate fifth generation language, every IoT devices eventually need engineer to code the algorithm into the program. Machine can’t just runs on its own, in order for things to get really “smart”, this definitely needs more creative human collaboration.

How IoT relates to big data and analytics?

Information consists of data, but data is not necessarily information. Also, wisdom is knowledge, which in turn is information, which in turn is data, but, for example, knowledge is not necessarily wisdom. So wisdom is a subset of knowledge, which is a subset of information, which is a subset of data (Steyn, 2001).

IoT reach out for things to monitor things, observe things to provide intelligence value, and facilitate users in massive data analytic. The concept of digital factory is a hype, on a fingertip, employees are able to drill down the time-based performance trend. For example, the daily clock in-out surely generate a huge amount of big data to the cloud. Massive data must be organized in a way (perhaps graphical chart) to provide informative presentation such as the absenteeism, or when is the estimated time of arrival of the truck based of the driver’s clock-out time. If user know how thing is going, how thing is feeling or their parameters and where things are, then they can better manage things. But here comes another challenging question, why does not the surveillance camera cannot perform the same job? The answer is eye judgment sometimes fails for accuracy, it is costly to allocate manpower to sit in front of the monitor and key in the clock-in time. With the system hooked up into the network, the data is automatically generated without fail.  

Big data analytics is the process of examining large data sets containing a variety of data types –i.e., big data – to uncover hidden patterns, unknown correlations, market trends, customer preferences and other useful business information. The analytical findings can lead to more effective marketing, new revenue opportunities, better customer service, improved operational efficiency, competitive advantages over rival organizations and other business benefits (Rouse 2010).

By sorting out the data, it provide more insight by analyzing a higher volume and variety of data types from more sources than ever before, by digging deeper into user information and behavior it enable marketing personalized based on individual preferences, gain faster insight by performing real-time analysis of user’s information to deliver offers at the point of decision. 

Challenges

What would be the challenges when implementing the IoT?

1. Data quality, the collected-data isn’t meet a desired standard, it happens a lot to the real world projects, where some desired data is not be able to be collected. To overcome this problem, BI (Business Intelligence) Consultant need to work hard in requirements gathering, more research on what attributes are critical to the client’s business.
2. Data analysis is not as simple as it seems, M2M is going to generate a massive amount of data, but with appropriate analytic to process the data to information does need a lot of expertise cost. Data mining is a technique to undermine the pattern in such massive amount of data, the challenge is how to store, analyze, predict and utilize the data being generated and collated.
3. Cost of buying sensors to install on the thing, the cost comprises of installation, maintenance, connectivity, and power.
4. Networks aren’t ubiquitous, the problem is interconnected networks aren’t available every corner in the world. In city area, we found ease of access to wireless networks, but not on outskirts city especially places with the population rate is low.
5. Security concern, the network segmentation need to be taken into account during implementation, especially when it comes to IoT for business and it involves sensitive data that can’t be expose to the outside world, network engineer need to isolate the network segment of the IoT device into a suitable subnet.
6. The sensor definitely need to go through a lot of improvise as underlying problems are yet to be discovered, there are numerous connection type invented such as NFC, iBeacon, infrared, Bluetooth, etc. Each has different limitation and its advantages, and it depends on the environment and interference, it is necessary to go through plenty of trial-and-error stages when during the implementation of IoT, cost is a burden for business.
7. Security is another concern, a security breach is any incident that results in unauthorized access of data by bypassing their underlying security mechanisms. The access control system sure generates a big amount of data, and the data is sensitive to closely affect the profit and loss of a business. The consequences of the security breach for any web provider will be the loss of revenue, damage to the reputation of the organization confidence, etc. In order to reduce overall risk, we can take the following methods to improve the level of network security for web operators, such as perimeter layer protection (Network Address Translation), provide secure website via HTTPS. 

Use cases in warehouse:

1.      Facial recognition control access. Figure 1 below shows the prototype of a biometric access control system, which primarily uses facial recognition to gain access to pass through the gate.

Figure 1 Facial Recognition Access Control

A facial recognition software has come a long way since 1970’s, but it was not publicly well accepted due to its stability to deliver the functionality. But it becomes popular nowadays due to lower cost of production and people are more concern in hygiene and convenience. The key idea is to link the software system to the internet, provides more value added functionality based on existing technology. It’s about how that data is used to extract valuable information and, subsequently, how that information is used to improve ‘something’. Subsequently the company RSH is going to expand the line of product from here onwards to products such as cloud Enterprise Resource Planning (ERP), smart scanner, smart watch for factory, etc.

2.      RFID tags, figure 2 shows the mechanism of a RFID tag. The primary functions of the RFID tag is aims to track the movement and storage of goods within a warehouse. When it pass through a gateway, it generates data. The warehouse management system therefore process the associated transactions.

Figure 2 RFID Tag

How it works? It uses and ADC (Automated Data Collection) technology that uses radio-frequency waves to transfer data between a reader and a movable item to identify, categorize, track. The technology benefits us by reducing the oversight possibility by human eyes. The RFID reader is a network connected device with an antenna that sends power as well as data and commands to the tags. Whereas, RFID tags (AKA transponder) consists of a chip, some memory and an antenna, it was attached to shipment item such as pallets, carton box, wooden cases, etc. With programmable unique ID number, we can set the item (e.g. pallet) an ID for endless possible uses.

There are two types of RFID, and they have different set of attributes:

	Active RFID	Passive RFID
Battery	Required	Not required
Signal strength between reader to tag	High	Very low
Tag power source	Internal to tag	Energy transferred from the reader via RF
Required signal strength from reader to tag	Very low	Very high (must power the tag)
Available of Tag Power	Continuous	Only within field of reader
Communication Range	Long range (100m or more)	Short or very short range (3m or less)
Sensor capability	Ability to continuously monitor and record sensor input; data/ time stamp for sensor events	Ability to read and transfer sensor values only when tag is powered by reader; no date/ time stamp
Data storage	Large read/ write data storage (128 bytes) with sophisticated data search and access capabilities available	Small read/ write data storage (e.g. 128 bytes)

Active vs Passive RFID (Source: http://www.cisco.com/)

Mode of connection:

1.      ANT	2.      Bluetooth	3.      DASH7
4.      HART	5.      ISA100	6.      Z-Wave
7.      Bluetooth	8.      Wireless HART	9.      Zigbee and Zigbee IP

Each connection type has different competitive advantage over the others, Bluetooth consumes less power but slower data rate of 723 kbps, whereas WiFi consumes more power relatively but with higher data rate 105 mbps. The choice of wireless radio technologies depends on the circumstances of how the device going to be used. The recommended connection type for a smart watch is Bluetooth since warehouse worker needs the mobility to move around without having to charge their device many times in a day, one charge can lasts for days.

How RFID does relates to IoT? Any object can attached with smart sensor and RFID tag to measure data. A warehouse can be remotely monitor via the wide area network, with RFID tag attached to an object, it generates data whenever it pass through a reader, user can see the movement and keeping track of its latest and previous locations. The reader identifies objects wirelessly without line-of-sight, the technology facilitates the warehouse management process.

The objective of implementing the IoT integrate with warehouse is to automate the operation, reduce losses and thefts in the storage.  Insurance company also can collaborate with Facial ID access control reseller, with client who implement Facial ID technology in their warehouse system, the insurance premium can be cheaper. 

Future of this technology

·         Data center infrastructure. The potential growth for this technology is limitless, but putting sensor and actuator on to devices or things, the usual thing becomes unusually valuable. E.g. pets, appliances, lights, cameras, home, cars, so on. Without human intervention, it automatically generate data to be stored on the cloud. Technology advances, population density increases, this definitely requires more and powerful data center to back up the rising demand.

·         IoT is an integrated part of the future, we can foresee the future world was scattered with IoT devices with IPv6, the physical and virtual things will be seamlessly integrated into the information network. It is certainly going to shape or change our living, running the business on anytime at anywhere through ubiquitous networking.

·         The growth of IoT seems to be limitless, because anything in the universe with a purpose of its creation, we can attach a sensor on it and connect it to the network. This gives the users more control for managing complex work processes.

·         There is a rising competition in the middleware solutions market, they provide the services to monitor and manage quality of service aspects, such as response time, resource consumption, throughput, availability, reliability and security. As competition get higher, the quality of communication will certainly improve over the time.