CSD operates on the same principle as a traditional voice call. When you initiated a CSD connection, the GSM network established a dedicated, unbroken, end-to-end circuit between your mobile device and the destination (for example, an Internet Service Provider's modem bank). This dedicated "pipe" was reserved exclusively for your data session for its entire duration, whether you were actively downloading a file or just reading a webpage. You were essentially making a "data call."

• Resource Allocation: The connection used one full-rate traffic channel (TCH/F), the same type of resource used for a voice call.
• Billing Model: Because a network resource was exclusively occupied, billing for CSD was based on time, just like a voice call. You were charged for every minute you were connected, regardless of how much data you actually transferred. This made it expensive for anything other than short, bursty data transfers.

The standard speed for a CSD connection was a mere $9.6 \text{ kbit/s}$ (kilobits per second). This was roughly one-fifth of the speed of a typical $56\text{k}$ dial-up modem at the time and is infinitesimally slow compared to modern 4G or 5G speeds. This limited its practical use to applications that required very little bandwidth, such as:

• Sending and receiving plain-text emails on a connected laptop.
• Sending faxes from a portable fax machine or laptop.
• Early remote access for industrial or telemetry applications (e.g., a remote sensor sending back a small data reading).
• Basic text-only web browsing using the Wireless Application Protocol (WAP).

To address the severe speed limitations of standard CSD, an enhancement called HSCSD was introduced. HSCSD was a clever way to boost data speeds by using a technique called channel bonding or multislot allocation.

Instead of assigning a user just one TDMA time slot, a network supporting HSCSD could allocate two, three, or even four time slots simultaneously to a single user's data session. This effectively multiplied the connection speed. For example:

• Using two time slots provided a speed of $19.2 \text{ kbit/s}$.
• Using four time slots could provide a speed up to $38.4 \text{ kbit/s}$.

While a significant improvement, HSCSD had its own set of drawbacks. Occupying multiple time slots meant you were using more network resources, so operators charged a premium for the service. Additionally, it required the phone's transmitter to be active for a much larger portion of each TDMA frame, leading to significantly increased power consumption and rapid battery drain. HSCSD was a niche service and ultimately a transitional technology, as the world was already moving towards more efficient packet-switched data.

The key to the success and elegance of SMS lies in how it is transported through the network. Unlike a voice call or a CSD session which require a dedicated traffic channel (TCH), SMS messages travel piggyback on the network's signaling channels.

Specifically, SMS messages are sent over the Standalone Dedicated Control Channel (SDCCH). This is the same channel used to handle call setup and location updates. Because this signaling channel is separate from the main traffic channels, it has several key advantages:

• You can send and receive text messages even while you are in the middle of a voice call.
• It utilizes otherwise unused capacity on the network's control pathways, making it very resource-efficient for the operator.

Another critical component of the SMS system is the Short Message Service Center (SMSC). The SMSC is a network database and server that functions on a "store-and-forward" basis. The process is as follows:

1. When you send a text message, your phone transmits it to your operator's SMSC.
2. The SMSC receives the message and queries the HLR to find out the location of the recipient's phone.
3. It then attempts to forward the message to the MSC/VLR where the recipient is currently located, which then delivers it to their handset.
4. If the recipient's phone is switched off or out of network coverage, the SMSC stores the message. It will periodically retry sending the message for a set period (e.g., 24-48 hours) until it is successfully delivered.

This store-and-forward mechanism makes SMS an extremely reliable form of communication, as messages are rarely lost even if the recipient is temporarily unavailable.

The iconic 160-character limit of an SMS message is a direct result of its design for efficiency on signaling channels. To keep messages small, GSM uses a special 7-bit character encoding set (the GSM 7-bit alphabet). The maximum size of the data payload for an SMS message is 140 octets (140 * 8 = 1120 bits). When using 7-bit characters, this space can accommodate $1120 / 7 = 160$ characters. For languages requiring characters outside this basic set (e.g., for characters with diacritics), a 16-bit Unicode encoding (UCS-2) is used, which reduces the limit to $1120 / 16 = 70$ characters per message segment.

• Calling Line Identification Presentation (CLIP) / Caller ID: This service presents the phone number of the person calling you on your phone's screen before you answer. The network retrieves the caller's number and includes it in the call setup signaling message sent to your device.
• Calling Line Identification Restriction (CLIR): This is the opposite of CLIP. It allows a caller to prevent their number from being displayed on the recipient's phone. This is a privacy feature activated by the caller before making a call, typically by dialing a special prefix.
• Call Waiting: This service notifies you of an incoming call while you are already on another call, usually with an audible beep. It gives you the option to place your current call on hold and answer the new call, or to ignore it.
• Call Hold: This allows you to temporarily pause an ongoing conversation without disconnecting it. You can put the other party on hold to, for example, answer a call on another line or to consult with someone privately before resuming the conversation.

This family of services allows you to automatically redirect incoming calls to another phone number (such as your voicemail, office phone, or another mobile number). GSM supports several conditions for forwarding:

• Call Forwarding Unconditional (CFU): All incoming calls are immediately redirected to the designated number without your phone ever ringing. Useful when you know you will be unavailable for an extended period.
• Call Forwarding on Busy (CFB): Incoming calls are redirected only when you are already on another call (your line is "busy").
• Call Forwarding on No Reply (CFNRy): If you do not answer an incoming call within a set amount of time (e.g., 20-30 seconds), it is automatically redirected. This is the mechanism commonly used to send unanswered calls to voicemail.
• Call Forwarding on Not Reachable (CFNRc): If an incoming call cannot be delivered to your phone because it is switched off or out of network coverage, it is redirected. This is another common trigger for voicemail.

• Multi-Party Service (Conference Calling): Allows a user to create a telephone conference by merging multiple incoming or outgoing calls into a single conversation.
• Barring of Calls: Allows a subscriber to block certain types of calls, such as barring all outgoing international calls to control costs, or barring all incoming calls when roaming.