Specifications

Sierra Wireless, Inc. CDPD Primer
2130006 Rev 1.0 Page 14
The signal could be weaker than desired in locations such as:
• Underground parking garages
• Tunnels
• Buildings with all metal construction
• Old concrete buildings with many steel reinforcing bars
In these cases, the CDPD subscriber may be able to get a stronger signal by locating the antenna
near an opening or window.
One of the functions of the CDPD service provider is to manage their CDPD network in order to
support a CDPD subscriber anywhere in their cellular geographic coverage area and deliver an
acceptable level of service. The CDPD service provider should also make the necessary
agreements with other providers so that roaming access is available outside of the original
provider’s coverage area.
3.2.7. Encryption and Security
Since CDPD is a public wireless data communications service that could be susceptible to
eavesdropping, all data transferred between the CDPD modem and the MD-IS (except broadcast
messages) is encrypted by CDPD’s Encryption Services, using RSA algorithms (see section
4.2.7). Data beyond the MD-IS is generally not encrypted, much as general Internet traffic
remains unencrypted unless the end user provides it.
3.2.8. Access Control and Congestion
Like wired Ethernet connections, CDPD is a contention-based system. It uses DSMA-CD
(digital sense, multiple access, collision detect), while Ethernet uses CSMA-CD (carrier sense,
multiple access, collision detect). In both systems, when a device has data to send it senses the
transmit medium to determine if it is currently busy (see section 5.5.1). If not, it will send its data,
and then wait to see if it is acknowledged by the receiver. If two devices did this at approximately
the same time a data “collision” would occur. The receiver would be unable to decode the
“smashed” data and would return a “decode failure” indicator. This indicator would be sensed by
the sending devices, which would then enter a random back-off period before trying again (see
section 5.5.2).
The CDPD-defined MAC (Medium Access Control) protocol is used to manage this function
(see section 5.5). Since each CDPD channel actually consists of physically separate forward
(network to modem) and reverse (modem to network) data paths, the system can operate in full
duplex mode. The forward (receive) channel maintains a busy/idle flag for the reverse (send)
channel, allowing the modem to monitor and determine when the channel is open.
The CDPD system allows a single user to access the radio channel data link while transmitting.
This inhibits all other users from transmitting until the currently active user becomes idle. The
maximum time a single user can access the radio channel on any single transmission is about
1 second. CDPD networks can allow a maximum of 64 blocks, of 385 bits each, in a single
transmission, but the actual limit is set by a parameter controlled by the MDBS. Initial access to
the channel is random, due to the contention-based nature of the access mechanism.
The impact of this mechanism to the CDPD user is that there may be small delays in accessing the
radio channel data link. When the user traffic on the channel is light, these delays are minor
(typically less than 0.1 seconds). When the traffic on the channel is heavy, the delays can become
larger (typically less than 1 second). As a result, user applications may experience slightly lower
apparent throughput when the CDPD radio channel data link becomes more heavily loaded. It is
fairly typical for the channel loading to vary cyclically throughout the day. Typically, channel
loading is light at midnight and heavier near daybreak, noon, and late afternoon.