WIKI

Ofon

OFON is a Public Switch Telephone Network (PSTN) in Indonesia that has official permission in accordance with Ministerial Decree PARPOSTEL No.47 / HK.501 / MPPT / 96 which was later changed to Ministerial Decree KOMINFO No.120 year 2015. Beside of being a pioneer in telephone service providers in Batam and Bintan islands, PT. Batam Bintan Telekomunikasi also held the official license from the Indonesian government as the third founder telecommunication company in Indonesia.

The history of OFON cannot be separated from the success of Industrial Estates in Batam and Bintan. At that time, PT Batam Bintan Telekomunikasi, together with other agencies from various fields, brought the Batam Area to become an international standard industrial area. This success has become the basic foundation for the development of OFON telephone services in which until now it can be used in  some territories in Indonesia. OFON which is based on IP services, will continue to grow and adapt to meet telecommunication needs with the latest technology and lifestyle.

1 History Ofon

OFON is a Public Switch Telephone Network (PSTN) in Indonesia that has official permission in accordance with Ministerial Decree PARPOSTEL No.47 / HK.501 / MPPT / 96 which was later changed to Ministerial Decree KOMINFO No.120 year 2015. Beside of being a pioneer in telephone service providers in Batam and Bintan islands, PT. Batam Bintan Telekomunikasi also held the official license from the Indonesian government as the third founder telecommunication company in Indonesia.

The history of OFON cannot be separated from the success of Industrial Estates in Batam and Bintan. At that time, PT Batam Bintan Telekomunikasi, together with other agencies from various fields, brought the Batam Area to become an international standard industrial area. This success has become the basic foundation for the development of OFON telephone services in which until now it can be used in  some territories in Indonesia. OFON which is based on IP services, will continue to grow and adapt to meet telecommunication needs with the latest technology and lifestyle.

2 OFON Products and Solutions

Codec is the process of encoding / decoding the voice source by a microphone in digital form to other participants in calls. There are several process included inside the Codec, such as Quantization, Compress and Decompression, and Encoding process. Before starting into the codec examples, we will touch a narrowband and wideband topics as it mainly used in the codec types.

2.1 Products

2.1.1 Home

A telephone service for modern homes with using SIP Phone for the communication. By a support of Internet, the SIP phone may be enabled using two extensions (landline & mobile). The Landline uses SIP phone with on premise solution while the mobile uses OFON IGO which is downloaded from google Playstore. The home offers a solution for employees whom unable to pick up phone from their houses.

2.1.2 Small Business

A telephone service for a small business scale. Similar to Home product, the small business offers a solution with 3 extensions (2 mobile phone & 1 landline). This product mainly used for Small Medium Enterprise (SME).

 

2.1.3 Lite Business

A telephone service for a large business scale. The lite business product may be integrated with the existing Analog PABX, that provides the Media Gateway to migrate the old telephone to the IP phone (SIP Phone). This telephone service provides 2, 4, and 8 capacities. Lite Business offers a solution to the customer whom like to maintain their existing PABX.

2.1.4 SIP Trunk

A telephone service for a large enterprise with a higher capacity supported more than 10 concurrent calls. The SIP Trunk configures on-premise IP address to enable routing calls inside the enterprise. SIP trunk offers a solution where customer prefer to maintain their own IP PABX on premise.

2.1.5 Smart PBX

A telephone service for a Small Medium Enterprise, Virtual Office, and Branch Office. The list of features in Smart PBX are IP PBX, Cloud Fax, and Conference Call. Smart PBX is mainly used for corporation services because of the flexibility and easier to manage by using a Web Portal.

2.2 Solutions

2.2.1 Proxy Number

A solution for bridging two calls by hiding their own private numbers to ensure the security between two users. This solution is also called as a Proxy Number, which aims to hide their numbers for calling their destination number. Proxy Number is a good solution for E-commerce to hide their identities to their customer using a business number. Figure 6 illustrates how the Proxy Number works.

2.2.2 Click2Call

A solution for enabling features in Android application to call the destination numbers. Similar to the proxy number, Click2Call is a good solution for E-commerce and Property Management System (PMS). This solution may be combined with Smart PBX or SIP Trunk to enable routing calls.

3 Two Types of Voice Network

Codec is the process of encoding / decoding the voice source by a microphone in digital form to other participants in calls. There are several process included inside the Codec, such as Quantization, Compress and Decompression, and Encoding process. Before starting into the codec examples, we will touch a narrowband and wideband topics as it mainly used in the codec types.

3.1 Traditional Voice Network

Traditional Voice Network, which is mainly based on TDM technology, uses fixed / dedicated line that guarantees the traffic voice. The term fixed line defines the guarantee of bandwidth in each channel when they have a communication session. If some calls are trying to reach these line, the calls will be queued, and will be wait for another line. The traditional voice is mostly called as Circuit Switched Network. The most common example of Circuit Switch Network is the PSTN and ISDN [1].

The main technology that used in circuit switch is using Plesiochronous Digital Hierarchy (PDH) and Synchronous Digital Hierarchy (SDH). The main differences between these two technologies are the synchronization between SDH and PDH. PDH is basically almost synchronous, because the bits in PD are stuffed into a frame as padding which means there is a slight variation between two frames. This might causing much delay if there are many bunches of voice traffic. In contrast, SDH allows a full complete synchronous which carries data each frame correctly.

Several standard units that commonly used in PDH are E1 (2.048 Mbps) , E2 (8.448 Mbps), E3 (34 Mbps), and E4 (140 Mbps). Each E1 consists of 32 channel slots, where 2 slots are used for signaling and frame synchronization, and the remain 30 slots are available. Each slot utilizes 64 Kbit/s, which is used for the voice call. Figure 7 illustrates the E1 concept with 30 reserved for telephone services and 2 slots for Frame Sync and Signaling.

Therefore, each slot is dedicated only for only one single call. If the other call is trying to reach the slot, this call should be queued or allocated to another slot.  PSTN ,  ISDN , and old traditional phones are using this technology with a guarantee bandwidth in each slot.

E1, E2, E3, and E4 are fundamentally the expansion of the voice call bandwidth (64 Kbps) which are multiplexed into a bunch of capacities. The E1 is multiplexed into E2, whereas E2 is multiplexed into E3, and so on. Figure 9 illustrates the evolution of E1, E2, E3, and E4.

 

The other technology besides of PDH is the SDH technology. SDH is a standard technology for synchronous data transmission on optical media [2], which brings more capacity compared to PDH in the network. The SDH utilizes Synchronous Transfer Module (STM) which has a hierarchy of multiplies 155.52 Mbit/s that is bigger than the PDH. The frame rate of STM is defined in ITU-T Recommendation G.707. The basic unit of SDH are:

  • STM-1                   : 155.52 Mbit/s
  • STM-4                  : 622.08 Mbit/s
  • STM-16                : 2488.32 Mbit/s
  • STM-64                : 9953.28 Mbit/s

The main advantages of using the SDH is the synchronization itself, where it may communicate to a lower systems without any synchronization. The PDH uses one slot of 64 Kbit/s in the E1 frame to synchronize the bit frame, this is why it called Plesiochronous. SDH is also compatible with any PDH configurations, such as E1, E2, and E3.

3.2 Modern Voice Network

Compared to the old traditional technology, the new type of voice network is based on IP technology. The IP itself delivers packet to the destination device by chopping several packets into different headers. The headers contain the destination packets and also the source address. This could be analogous by using the concept of postman letter. The letter consists of the recipient and the destination address, which make it easily for the post office to identify the destination place. A same concept of the IP packet, the postman itself is the IP protocol whereas destination address is written inside the header. The letter refers to the packet. This technique of inspection and retransmission is called store-and-forward.

By using the concept of the postman, this type of voice network is using a shared line without voice traffic guarantee. This packets do not guarantee the type of traffics if there is any delay or latency in the network, since one line is used for several type of packets (i.e: synchronization, voice, video, text messaging).

To send the packet from the source to the destination, the Transport Layer in Layer 4 defines two type of connections, which are connectionless and connection-oriented. The Connection-Oriented guarantees packet from the source to the destination using three-ways handshakes. In contrast, the connectionless type does not guarantee any packets through the destination which mainly treated independently. With the VoIP perspective, the connectionless is mostly popularized by Internet Protocol (IP) in VoIP system.  

4 How VoIP Work?

Codec is the process of encoding / decoding the voice source by a microphone in digital form to other participants in calls. There are several process included inside the Codec, such as Quantization, Compress and Decompression, and Encoding process. Before starting into the codec examples, we will touch a narrowband and wideband topics as it mainly used in the codec types.

4.1 Signaling

The main function of signaling is to give the first notification from the source to the destination number by giving a ring call. This signaling is required mainly to allocate the IP address of the destination number by sending them a notification. In fact, most destination number locations are located in random places, which hardly to track them. This is one of the main function of signaling that allows to find the IP address by surfing the Internet to identify the destination IP address.

In the concept of signaling, there are two protocols that mostly used to maintain a call session, which are H.323 and Session Initiation Protocol (SIP). In fact, the H.323 currently replacing SIP because some features in H.323 are mostly a part of the SIP features. SIP itself is one of the technologies adapted from IETF that aims to replace the H.323 with more advanced features, such as creating, modification, and terminating multimedia and calls.

The basic principle of SIP is using three way handshake to secure the signaling conversation. For example, Bob and Alice are trying to have conversation using VoIP. Several steps that used by both users to communicate are:

  • Bob INVITE Alice in the conversation
  • Alice notifies Bob by saying “YES” which means the conversation is accepted by Alice
  • Bob acknowledge (ACK) Alice by giving a route SIP.
  • The conversation occurs
  • The conversation terminates (Bob terminates using BYE)

Figure 11 Three Way Handshake conversation in SIP [3]

4.2 Communicating between source and destination

Compared to traditional phone, which is using the dedicated line to transmit the voice signal that allows only one call at a same time to transmit in the same dedicated line, the shared line enable to have multiple calls at a same time in one line. This lead to an efficient way for saving Total Cost of Ownership (TCO). Therefore, VoIP is a future technology to bring a voice signal into the end user.

In the VoIP transmission, to communicate between two users, there are five different steps to send a traffic signal from a source to the destination :

  • Analog to Digital Conversion
  • Digital Signal Encapsulation into Voice over IP (VoIP)
  • Internet
  • VoIP Decapsulation into Analog
  • Digital to Analog Conversion

 Figure 10 illustrates the encapsulation process of a voice to the Internet Protocol (IP).

Figure 12 Encapsulation Process in Voice over IP (VoIP)

 

4.2.1 Analog to Digital Conversion

The first process in the VoIP packet is to convert the Analog to Digital voice. This process is called Analog to Digital Conversion (ADC) which is occurred in the Source Number / A Number as illustrated in Figure 10. The ADC concept uses three different steps :

  • Sampling
  • Quantisation
  • Coding

4.2.1.1 Sampling

The sampling process is the technique to convert the continuous signal (analog) to a discrete signal (digital) by taking a different sample of time signal. The main important of the sampling process is the Time Sampling, which defines the duration sampling of the continuous signal. If the time sampling rate is high, the total quantity of samples will be lower. In contrast, if the time sampling is relatively low, the total quantity of samples will be higher.

Figure 13 The Sampling Rate from (A : Continuous Signal ) to (B : Discrete Signal )

Table 1 Time Sampling vs Total Quantity of Sampling

There are such a correlation between time sampling, total quantity of samples, accuracy and time process, which leads to a trade off between each signal. The time sampling basically affects the accuracy of the signal; if the time sampling is higher, the accuracy of data signal converted from analog to digital will be inaccurate. In contrast, as the time sample becomes lower, which means the total quantity of samples is larger, will lead to an accurate signal.

This accuracy is also caused by the total quantity of samples. This samples reflect the time process of each device to convert the analog to digital quickly, where larger quantity of samples lead to more accurate signal. However, larger number of samples needs a longer time process for device to accomplish the work. The trade off among all variables mentioned in Table 1 leaves a choice to different type of users. 

4.2.1.2 Quantization

The quantization is the second process in ADC to convert the sampled data to a quantized value that readable for the user. The quantization formula is using , where N is equal to 1,2,3… N, depends on the sample rate chosen in sampling process. Therefore, the expected result for each quantization would be multiplied by 2, such as 2, 4, 8, 16, 32 … etc. If the sampled data didn’t achieve this value, the sampled data may be ceiling up or down. Table 2 illustrates the sampled data vs quantization signal example cases and Figure 12 illustrates the quantization value in the discrete signal after being obtaining the quantization signal. 

Table 1 Sampled data  vs Quantization Signal

Figure 14 Quantization value in the discrete signal

4.2.1.3 Coding

The coding process is the final process in ADC conversion, where it aims to convert the decimal number of quantized signal to the binary numbers. Since we obtained a numeric number (2, 4, 8, 16) from the quantized signal, the numeric number can be converted as follows:

Figure 15 Numeric Numbers to Digital Numbers

4.2.2 Encapsulation of Digital to Voice over IP

Once the analog signal has been converted to the digital signal, the next step is the encapsulation of voice signal into IP packet. The most important protocols are Real Time Protocol (RTP) and User Datagram Protocol (UDP) which are the protocols to run voice signal into the IP packet. The UDP is chosen as the Layer 4 protocol to transmit voice signal since its support connectionless, which means by sending a data using UDP, there is no correction code or guarantee of loss.  Therefore, by using UDP, the packet is keep sending even though error occurs.

Related to RTP, the RTP is mainly used to monitor congestion problem such as packet loss, high latency, and bandwidth monitoring. RTP alongside with SIP will work together to send voice and video from the source.

Figure 4 illustrates the voice signal that encapsulated into the IP packet. OFON mostly used G.711 and G.729 for the codec, which requires 160 and 20 bytes respectively.

Figure 16 Voice Packet Encapsulation in the IP Packet

4.2.3 Voice over Internet Protocol

As illustrated in Figure 3, the Voice has been transmitted into the IP protocol using UDP and RTP headers. The voice signal also has been digitized so we expect to have binary code in this packet. This packet will be sent to the destination number through different types of connection. Corporates or offices may use Metro Ethernet (Metro-E), PPP, ATM, HDLC, or MPLS Network which really depends on their preferences. The essential point is the voice signal could identify the next destination number by identifying the routing table in each router. This is the way they communicate each other.

4.2.5 Decapsulation of VoIP into Digital

After the voice packet arrives in the destination address, the voice packet should be decapsulated from the IP Headers. The decapsulation of VoIP into digital is using the same concept as the encapsulation. The main differences is only Voice Packet is decapsulated from the IP headers as illustrated in Figure 15. The digitized voice is separated from the IP headers.

Figure 17 Voice Packet Decapsulation in the IP

4.2.5 Digital to Analog Conversion (DAC)

Digital to Analog Conversion has the same concept as ADC. The main process of DAC is a reverse of ADC, which includes the coding, quantization, and sampling rate. Therefore, the first process is to decode the binary numbers (1000, 0100) into decimal numbers (8, 4). Afterwards, the decimal numbers who hold the quantisation value will be decoded into the analog signal. Thus, we obtained the analog signal correctly.

5 Codec

Codec is the process of encoding / decoding the voice source by a microphone in digital form to other participants in calls. There are several process included inside the Codec, such as Quantization, Compress and Decompression, and Encoding process. Before starting into the codec examples, we will touch a narrowband and wideband topics as it mainly used in the codec types.

5.1 Wideband vs Narrowband

Both bands are working well in the voice communication. However, Wideband offers a larger bandwidth and a faster communication, but it needs larger power to transmit in the channel. As it needs more power, the wideband frequency is more vulnerable to noise compared to Narrowband. 

In contrary, Narrowband uses a small bandwidth and a slower communication. However, since narrowband uses small power to transmit in the channel [4][5], the noise in narrowband is relatively lower compared to Wideband. In conclusion, both hands have their own advantages and disadvantages, depends on the user experience who need a clear channel or either a clear noise. Table 2 describes the main differences between Wideband and Narrowband.

Table 2 Overview Wideband vs Narrowband

 

Wideband and Narrowband are mostly used in different types of codes. The type of codes are differentiated as the topic as follows [6].  

5.2 G.711

The G.711 utilizes 64 Kbit/s analog voice converted to digital voice with using a Narrowband Frequency. The G.711 is mainly divided with two types, which are G.711 A-law and U-law [7]. The main differences between A-Law and U-Law are the dynamic range provided by Law.

A-Law refers to the European PCM Systems. The main concept of A-Law is to use a linear segment for small amplitudes only, whereas logarithmic segment is mainly used for larger amplitudes. Figure 17 illustrates the concept of A-Law. It shows that in a small amplitudes, it is using a linear graph compared to logarithmic. This circumstance leads to a A-Law compression [7][8].

Figure 18 The A-Law Concept

Because of this linearity, the A-Law has a smaller dynamic range compared to U-Law. This Dynamic Range is mainly used to identify the ratio between the loudest and quietest in the signal. The higher variant of ranges in the signal, the higher distortion would be. Therefore, the A-Law will have a better sound compared to U-Law [7][8].

The U-Law refers to the American PCM Systems with a larger dynamic range compared to A-law. In contrast to A-Law, U-Law uses logarithmic segment in all values. This leads to a higher dynamic range between the quietest and loudest sound which represented by the signal. Figure 7 below shows the concept of U-Law.

5.3 G.729AB

The G.729AB codes utilizes the lower bandwidth with 8 Kbit/s, with a Wideband range. In contrast, with G.711, this codec is mostly used in the Low Bandwidth Requirement since G.729 will decrease the speech quality significantly [9]. The term of AB in this codec is the extension of G.729, where A defines less computational power and annex B defines a silence compression method that enable Voice Activity Detection (VAD). VAD is mainly used to detect voice in the background noise. In term of AB, the advantages of annex A and B are combined.

5.4 G.722

As a development of G.711, G.722 employs a wideband audio codec that operating in 48 and 64 Kbit/s to improve the speech quality of narrowband in G.711. Since the G.711 using a bandwidth of human speech 300 – 3400 Hz, this codec increase the bandwidth into 50 – 7000 Hz. However, G.722 allows more bandwidth into the channel that potentially more vulnerable to noise compared to G.711 [10].

5.5 OPUS

Opus is one of the codec that mainly used in the mobile communication system with a variant location. It can scale from a low bitrate to a very high quality stereo music, which covers Narrowband, Wideband, Full band. The working frequency is ranged from 6 to 510 Kbit/s , with a sampling rate 8 KHz (narrowband) to 48 KHz (full band). OPUS combined a G.711 and G.722 features to cater the ranges between low and high qualities.

[1]

S. Kashihara, VoIP Technologies. 2011.

[2]

Prasanta Kumar Sahu, “SDH and PDH Technologies,” Ericsson, 2010.

[3]

中島一憲, No Titleメンタルヘルスハンドブック教師のストレス総チェック. 2000.

[4]

Stack Exchange, “rf - Wide band vs Narrow band communication - Electrical Engineering Stack Exchange.” [Online]. Available: https://electronics.stackexchange.com/questions/158069/wide-band-vs-narrow-band-communication. [Accessed: 10-Dec-2018].

[5]

Microwaves & RF, “Comparing Narrowband and Wideband Channels.” [Online]. Available: https://www.mwrf.com/systems/comparing-narrowband-and-wideband-channels. [Accessed: 10-Dec-2018].

[6]

TechTarget, “What is codec? - Definition from WhatIs.com,” 2017. [Online]. Available: https://searchunifiedcommunications.techtarget.com/definition/codec. [Accessed: 29-Nov-2018].

[7]

D. M. Rodrigues, “Introduction to Telecommunication Network - Lecture Notes,” 2016.

[8]

D. Between.net, “Difference Between A-law and u-Law,” Between.net, Difference. [Online]. Available: http://www.differencebetween.net/technology/difference-between-a-law-and-u-law/.

[9]

Speech Codecs, “Understanding Various Speech Codecs,” 2008. [Online]. Available: https://speechcodecs.wordpress.com/tag/g729ab/.

[10]

TechTarget, “What is G.722? - Definition from WhatIs.com,” TechTarget. [Online]. Available: https://searchunifiedcommunications.techtarget.com/definition/G722. [Accessed: 11-Jan-2019].